diff --git a/.env.example b/.env.example
index 228d697..7cc4641 100644
--- a/.env.example
+++ b/.env.example
@@ -1,25 +1,25 @@
-# LLM Provider Configuration
-# ===========================
+# ══════════════════════════════════════════════════════════════
+# REQUIRED: At least ONE LLM provider key
+# ══════════════════════════════════════════════════════════════
+OPENAI_API_KEY=sk-...
+# ANTHROPIC_API_KEY=
+# GEMINI_API_KEY=
+# DEEPSEEK_API_KEY=
 
-# Default provider to use (openai, anthropic, gemini)
-DEFAULT_LLM_CLIENT=openai
+# ══════════════════════════════════════════════════════════════
+# REQUIRED for visual element detection (Agent.Find Visual Element)
+# ══════════════════════════════════════════════════════════════
+HUGGINGFACE_API_KEY=hf_...
 
-# OpenAI Configuration
-OPENAI_API_KEY=your_openai_api_key_here
-DEFAULT_OPENAI_MODEL=gpt-4o-mini
-
-# Anthropic Configuration
-ANTHROPIC_API_KEY=your_anthropic_api_key_here
-DEFAULT_ANTHROPIC_MODEL=claude-sonnet-4-5-20250929
-
-# Google Gemini Configuration
-GEMINI_API_KEY=your_gemini_api_key_here
-DEFAULT_GEMINI_MODEL=gemini-1.5-flash
-
-# Image Upload Services
-IMGBB_API_KEY=your_imgbb_api_key_here
-FREEIMAGEHOST_API_KEY=your_freeimagehost_api_key_here
-
-# HuggingFace (Future)
-HUGGINGFACE_API_KEY=your_huggingface_api_key_here
+# ══════════════════════════════════════════════════════════════
+# OPTIONAL: Customize defaults
+# ══════════════════════════════════════════════════════════════
+# DEFAULT_LLM_CLIENT=openai
+# DEFAULT_OPENAI_MODEL=gpt-4o
+# OLLAMA_BASE_URL=http://localhost:11434/v1
 
+# ══════════════════════════════════════════════════════════════
+# OPTIONAL: Image upload (falls back to base64 if not set)
+# ══════════════════════════════════════════════════════════════
+# IMGBB_API_KEY=
+# FREEIMAGEHOST_API_KEY=
\ No newline at end of file
diff --git a/Agent/AgentKeywords.py b/Agent/AgentKeywords.py
index c5c50b1..25471bf 100644
--- a/Agent/AgentKeywords.py
+++ b/Agent/AgentKeywords.py
@@ -23,18 +23,20 @@ class AgentKeywords:
 
     def __init__(
         self, 
-        llm_client: str = "openai", 
-        llm_model: str = "gpt-4o-mini",
+        llm_client: str | None = None, 
+        llm_model: str | None = None,
         platform_type: str = "auto",
-        click_mode: str = "xml",
-        input_mode: str = "text",
+        element_source: str = "accessibility",
+        llm_input_format: str = "som",
     ):
+        from Agent.config.config import Config
+        
         self.engine = AgentEngine(
-            llm_client=llm_client, 
+            llm_client=llm_client or Config.DEFAULT_LLM_CLIENT, 
             llm_model=llm_model,
             platform_type=platform_type,
-            click_mode=click_mode,
-            input_mode=input_mode,
+            element_source=element_source,
+            llm_input_format=llm_input_format,
         )
 
     # ----------------------- Public RF Keywords -----------------------
@@ -64,11 +66,23 @@ def find_visual_element(self, description: str, format: str = "center"):
         """
         return self.engine.find_visual_element(description, format=format)
 
-    def autonumous(self, instruction: str):
-        """Agent.Autonumous <instruction>
-        This keyword is designed to autonomously plan and execute a test based on the 
-        given single instruction.
+    def set_element_source(self, source: str):
+        """Agent.Set Element Source    accessibility|vision
+        Example: Agent.Set Element Source    vision
+        """
+        self.engine.set_element_source(source)
+
+    # def set_llm_input_format(self, format: str):
+    #     """Agent.Set LLM Input Format    text|som
+    #     Example: Agent.Set LLM Input Format    som
+    #     """
+    #     self.engine.set_llm_input_format(format)
+
+    # def autonumous(self, instruction: str):
+    #     """Agent.Autonumous <instruction>
+    #     This keyword is designed to autonomously plan and execute a test based on the 
+    #     given single instruction.
 
-        Example: Agent.Autonumous    Navigate to settings, change language to French, 
-        ...    then go back to home screen and verify the interface is in French"""
-        raise NotImplementedError("Agent.Autonumous is not implemented yet")
\ No newline at end of file
+    #     Example: Agent.Autonumous    Navigate to settings, change language to French, 
+    #     ...    then go back to home screen and verify the interface is in French"""
+    #     raise NotImplementedError("Agent.Autonumous is not implemented yet")
\ No newline at end of file
diff --git a/Agent/__init__.py b/Agent/__init__.py
index ae3af54..137da1b 100644
--- a/Agent/__init__.py
+++ b/Agent/__init__.py
@@ -1,7 +1,7 @@
 
 from Agent.AgentKeywords import AgentKeywords
 
-__version__ = "0.1.0"
+__version__ = "0.0.1"
 __all__ = ["AgentKeywords"]
 
 
diff --git a/Agent/agent_engine.py b/Agent/agent_engine.py
index 48b5702..dbba091 100644
--- a/Agent/agent_engine.py
+++ b/Agent/agent_engine.py
@@ -1,6 +1,6 @@
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Dict, List, Optional
 
-from Agent.platforms import DeviceConnector, WebConnectorRF, create_platform
+from Agent.platforms import DeviceConnector, create_platform
 from Agent.ai.llm.facade import UnifiedLLMFacade
 from Agent.ai._promptcomposer import AgentPromptComposer
 from Agent.utilities.imguploader.imghandler import ImageUploader
@@ -8,114 +8,84 @@
 from Agent.tools.base import ToolCategory
 from Agent.core.keyword_runner import KeywordRunner
 from Agent.tools.mobile import MOBILE_TOOLS
-from Agent.tools.web import WEB_TOOLS
 from Agent.tools.visual import VISUAL_TOOLS
 from robot.api import logger
 
 
 class AgentEngine:
-    """Core engine for AI-driven test automation.
-    
-    Orchestrates the complete Agent.Do and Agent.VisualCheck flows:
-    - Capturing UI context and screenshots
-    - Composing AI prompts and calling LLMs
-    - Tool-based action execution and visual verification
-    
-    This is the main orchestrator that coordinates platform connectors,
-    AI services, tool registry, and executors.
-    """
+    """Core engine for AI-driven Android test automation."""
 
     def __init__(
         self, 
         llm_client: str = "openai", 
         llm_model: str = "gpt-4o-mini",
-        platform: Optional[Union[DeviceConnector, WebConnectorRF]] = None,
+        platform: Optional[DeviceConnector] = None,
         platform_type: str = "auto",
-        click_mode: str = "xml",
-        input_mode: str = "text",
+        element_source: str = "accessibility",
+        llm_input_format: str = "text",
     ) -> None:
-        # Platform connector - create or use provided
         if platform is None:
             self.platform = create_platform(platform_type)
         else:
             self.platform = platform
         
-        # Detect platform type
-        platform_name = self.platform.get_platform()
-        logger.info(f"🌐 Platform detected: {platform_name}")
+        logger.info("📱 Platform: mobile")
         
-        # AI components
         self.llm = UnifiedLLMFacade(provider=llm_client, model=llm_model)
-        self.prompt_composer = AgentPromptComposer(
-            platform_connector=self.platform
-        )
         self.image_uploader = ImageUploader(service="auto")
         
-        # Tool execution components
         self.tool_registry = ToolRegistry()
         self.executor = KeywordRunner(self.platform)
         
-        # Register tools based on platform
-        if platform_name == "web":
-            self._register_web_tools()
-        else:
-            self._register_mobile_tools()
+        self._register_mobile_tools()
         self._register_visual_tools()
         
-        # Click strategy and input mode
-        self.click_mode = click_mode
-        self.input_mode = input_mode
-        logger.info(f"🎯 Click mode: {click_mode}, Input mode: {input_mode}")
+        self.prompt_composer = AgentPromptComposer(
+            tool_registry=self.tool_registry,
+            platform_connector=self.platform
+        )
+        
+        self.element_source = element_source
+        self.llm_input_format = llm_input_format
+        logger.info(f"🎯 Element source: {element_source}, LLM input format: {llm_input_format}")
     
     def _register_mobile_tools(self) -> None:
-        """Register all mobile tools in the registry."""
         for ToolClass in MOBILE_TOOLS:
             self.tool_registry.register(ToolClass())
-        
         mobile_tools_count = len(self.tool_registry.get_by_category(ToolCategory.MOBILE))
         logger.debug(f"📱 Registered {mobile_tools_count} mobile tools")
     
-    def _register_web_tools(self) -> None:
-        """Register all web tools in the registry."""
-        for ToolClass in WEB_TOOLS:
-            self.tool_registry.register(ToolClass())
-        
-        web_tools_count = len(self.tool_registry.get_by_category(ToolCategory.WEB))
-        logger.debug(f"🌐 Registered {web_tools_count} web tools")
-    
     def _register_visual_tools(self) -> None:
-        """Register all visual verification tools in the registry."""
         for ToolClass in VISUAL_TOOLS:
             self.tool_registry.register(ToolClass())
-        
         visual_tools_count = len(self.tool_registry.get_by_category(ToolCategory.VISUAL))
         logger.debug(f"👁️ Registered {visual_tools_count} visual tools")
     
     # ----------------------- Public API -----------------------
     
-    def set_click_mode(self, mode: str) -> None:
-        """Change click mode dynamically during test execution.
+    def set_element_source(self, source: str) -> None:
+        """Change element source dynamically.
         
         Args:
-            mode: 'xml' or 'visual'
+            source: 'accessibility' or 'vision'
         """
-        if mode not in ["xml", "visual"]:
-            raise ValueError(f"Invalid click_mode: {mode}. Choose: xml, visual")
+        if source not in ["accessibility", "vision"]:
+            raise ValueError(f"Invalid element_source: {source}. Choose: accessibility, vision")
         
-        self.click_mode = mode
-        logger.info(f"🔧 Click mode changed to: {mode}")
+        self.element_source = source
+        logger.info(f"🔧 Element source changed to: {source}")
     
-    def set_input_mode(self, mode: str) -> None:
-        """Change input mode dynamically during test execution.
+    def set_llm_input_format(self, format: str) -> None:
+        """Change LLM input format dynamically.
         
         Args:
-            mode: 'text' (numbered list) or 'som' (screenshot with numbered boxes)
+            format: 'text' or 'som'
         """
-        if mode not in ["text", "som"]:
-            raise ValueError(f"Invalid input_mode: {mode}. Choose: text, som")
+        if format not in ["text", "som"]:
+            raise ValueError(f"Invalid llm_input_format: {format}. Choose: text, som")
         
-        self.input_mode = mode
-        logger.info(f"🔧 Input mode changed to: {mode}")
+        self.llm_input_format = format
+        logger.info(f"🔧 LLM input format changed to: {format}")
     
     def do(self, instruction: str) -> None:
         """Execute AI-driven action based on natural language instruction.
@@ -125,21 +95,64 @@ def do(self, instruction: str) -> None:
         """
         logger.info(f"🚀 Starting Agent.Do: '{instruction}'")
 
-        # Collect UI context (skip in visual-only mode)
+        if hasattr(self.platform, 'wait_for_page_stable'):
+            self.platform.wait_for_page_stable()
+
+        screenshot_base64 = None
         ui_candidates = []
-        if self.click_mode != "visual":
-            ui_candidates = self.platform.collect_ui_candidates()
-        else:
-            logger.debug("⚡ UI collection skipped (mode: visual)")
+        annotated_image_path = None
         
-        # Capture screenshot if needed (for SoM mode or visual click mode)
-        screenshot_base64 = None
-        need_screenshot = self.input_mode == "som" or self.click_mode == "visual"
-        if need_screenshot:
+        # Collect UI elements based on element source
+        if self.element_source == "accessibility":
+            ui_candidates = self.platform.collect_ui_candidates()
+            logger.debug(f"📋 Collected {len(ui_candidates)} accessibility elements")
+        elif self.element_source == "vision":
             screenshot_base64 = self.platform.get_screenshot_base64()
-            logger.debug(f"📸 Screenshot captured (input_mode: {self.input_mode})")
-        else:
-            logger.debug(f"⚡ Screenshot skipped (input_mode: {self.input_mode})")
+            from Agent.ai.vlm._client import OmniParserClient
+            from Agent.ai.vlm._parser import OmniParserResultProcessor
+            from PIL import Image
+            
+            client = OmniParserClient()
+            image_temp_path, parsed_text = client.parse_image(image_base64=screenshot_base64)
+            annotated_image_path = image_temp_path
+            
+            if parsed_text:
+                processor = OmniParserResultProcessor(
+                    response_text=parsed_text,
+                    image_temp_path=image_temp_path,
+                )
+                elements_data = processor.get_parsed_ui_elements(element_type="all")
+                
+                with Image.open(image_temp_path) as img:
+                    width, height = img.size
+                
+                for key, data in elements_data.items():
+                    bbox_norm = data.get("bbox", [0, 0, 0, 0])
+                    x1 = int(bbox_norm[0] * width)
+                    y1 = int(bbox_norm[1] * height)
+                    x2 = int(bbox_norm[2] * width)
+                    y2 = int(bbox_norm[3] * height)
+                    
+                    element = {
+                        "text": data.get("content", ""),
+                        "class_name": data.get("type", "unknown"),
+                        "bbox": {
+                            "x": x1,
+                            "y": y1,
+                            "width": x2 - x1,
+                            "height": y2 - y1
+                        },
+                        "source": "omniparser",
+                        "interactivity": data.get("interactivity", "unknown")
+                    }
+                    ui_candidates.append(element)
+                
+            logger.debug(f"👁️ Detected {len(ui_candidates)} visual elements")
+        
+        # Capture screenshot if needed for SoM mode and not already captured
+        if self.llm_input_format == "som" and not screenshot_base64:
+            screenshot_base64 = self.platform.get_screenshot_base64()
+            logger.debug("📸 Screenshot captured for SoM mode")
         
         # Prepare context for tool execution
         context = {
@@ -150,17 +163,28 @@ def do(self, instruction: str) -> None:
         if screenshot_base64:
             context["screenshot_base64"] = screenshot_base64
         
+        logger.info(f"Elements sent to AI: {ui_candidates}")
+        
         # Prepare AI request
         platform_name = self.platform.get_platform()
+        tool_category = "mobile"
         messages = self.prompt_composer.compose_do_messages(
             instruction=instruction,
             ui_elements=ui_candidates,
             platform=platform_name,
-            click_mode=self.click_mode,
-            input_mode=self.input_mode,
+            element_source=self.element_source,
+            llm_input_format=self.llm_input_format,
             screenshot_base64=screenshot_base64,
+            annotated_image_path=annotated_image_path,
         )
-        tools = self.prompt_composer.get_do_tools(category=platform_name, click_mode=self.click_mode)
+        if annotated_image_path:
+            logger.info(f"Annotated image: {annotated_image_path}")
+        
+        tools = self.prompt_composer.get_do_tools(category=tool_category, element_source=self.element_source)
+        logger.debug(f"Tools for {tool_category}: {len(tools)} tools")
+        
+        if not tools:
+            raise RuntimeError(f"No tools registered for platform '{platform_name}'. Check tool registration.")
         
         # Call AI
         result = self.llm.send_ai_request_with_tools(
@@ -169,12 +193,15 @@ def do(self, instruction: str) -> None:
             tool_choice="required",
             temperature=0
         )
-
-        logger.debug(f"AI response: {result}")
+        
+        tool_call = result.get("tool_calls", [{}])[0]
+        tool_name = tool_call.get("function", {}).get("name", "unknown")
+        tool_args = tool_call.get("function", {}).get("arguments", {})
+        logger.info(f"🤖 AI chose: {tool_name}({tool_args})")
         
         # Execute tool
         self._execute_do_from_tool_calls(result, context, instruction)
-        logger.info("✅ Agent.Do completed")
+        logger.info("Agent.Do completed")
 
     def visual_check(self, instruction: str) -> None:
         """Execute visual verification based on natural language instruction.
@@ -184,6 +211,10 @@ def visual_check(self, instruction: str) -> None:
                         (e.g., "verify the home screen is displayed")
         """
         logger.info(f"👁️ Starting Agent.VisualCheck: '{instruction}'")
+
+        if hasattr(self.platform, 'wait_for_page_stable'):
+            self.platform.wait_for_page_stable()
+
         screenshot_base64 = self.platform.get_screenshot_base64()
         
         # Embed screenshot to Robot Framework log
@@ -194,6 +225,10 @@ def visual_check(self, instruction: str) -> None:
         # Prepare AI request
         messages = self.prompt_composer.compose_visual_check_messages(instruction, image_url)
         tools = self.prompt_composer.get_visual_check_tools()
+        logger.debug(f"Visual check tools: {len(tools)} tools")
+        
+        if not tools:
+            raise RuntimeError("No visual tools registered. Check tool registration.")
         
         # Call AI
         result = self.llm.send_ai_request_with_tools(
@@ -219,6 +254,10 @@ def ask(self, question: str, response_format: str = "text") -> str:
         """
         import json
         logger.info(f"❓ Agent.Ask: '{question}'")
+        
+        if hasattr(self.platform, 'wait_for_page_stable'):
+            self.platform.wait_for_page_stable()
+        
         screenshot_base64 = self.platform.get_screenshot_base64()
         self.platform.embed_image_to_log(screenshot_base64)
         
@@ -248,6 +287,10 @@ def find_visual_element(self, description: str, format: str = "center") -> Dict[
         from Agent.ai.vlm.interface import OmniParserOrchestrator
         
         logger.info(f"🔍 Agent.Find Visual Element: '{description}'")
+        
+        if hasattr(self.platform, 'wait_for_page_stable'):
+            self.platform.wait_for_page_stable()
+        
         screenshot_base64 = self.platform.get_screenshot_base64()
         self.platform.embed_image_to_log(screenshot_base64)
         
@@ -259,7 +302,7 @@ def find_visual_element(self, description: str, format: str = "center") -> Dict[
         result = orchestrator.find_element(
             element_description=description,
             image_base64=screenshot_base64,
-            element_type="interactive"
+            element_type="all"
         )
         
         if not result:
diff --git a/Agent/ai/_promptcomposer.py b/Agent/ai/_promptcomposer.py
index 20834eb..df3324c 100644
--- a/Agent/ai/_promptcomposer.py
+++ b/Agent/ai/_promptcomposer.py
@@ -1,6 +1,10 @@
 from typing import List, Dict, Optional, Any
 from Agent.tools.registry import ToolRegistry
 from Agent.tools.base import ToolCategory
+from robot.api import logger
+import base64
+import os
+from datetime import datetime
 
 
 class AgentPromptComposer:
@@ -13,25 +17,46 @@ def __init__(
     ) -> None:
         self.registry = tool_registry or ToolRegistry()
         self.platform = platform_connector
+        self._annotated_dir = None
+    
+    def _get_annotated_dir(self) -> str:
+        if self._annotated_dir is None:
+            from Agent.utilities._logdir import set_artifacts_subdir
+            self._annotated_dir = set_artifacts_subdir("RF_Agent/Annotated")
+        return self._annotated_dir
+    
+    def _save_annotated_image(self, image_base64: str, source: str = "som") -> str:
+        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S_%f")[:-3]
+        filename = f"annotated_{source}_{timestamp}.png"
+        filepath = os.path.join(self._get_annotated_dir(), filename)
+        
+        image_bytes = base64.b64decode(image_base64)
+        with open(filepath, "wb") as f:
+            f.write(image_bytes)
+        
+        logger.info(f"📸 Saved annotated image: {filepath}")
+        return filepath
 
     def compose_do_messages(
         self,
         instruction: str,
         ui_elements: Optional[List[Dict[str, Any]]] = None,
         platform: str = "mobile",
-        click_mode: str = "xml",
-        input_mode: str = "text",
+        element_source: str = "accessibility",
+        llm_input_format: str = "text",
         screenshot_base64: Optional[str] = None,
+        annotated_image_path: Optional[str] = None,
     ) -> List[Dict[str, Any]]:
         """Build DO action messages using tool calling approach.
         
         Args:
             instruction: User instruction
             ui_elements: List of UI elements
-            platform: 'mobile' or 'web' - determines system prompt
-            click_mode: 'xml' or 'visual' - guides AI on click strategy
-            input_mode: 'text' (numbered list) or 'som' (screenshot with numbered boxes)
-            screenshot_base64: Screenshot for SoM mode (required if input_mode='som')
+            platform: 'android' or 'ios'
+            element_source: 'accessibility' or 'vision'
+            llm_input_format: 'text' or 'som'
+            screenshot_base64: Screenshot (required for SoM mode)
+            annotated_image_path: Pre-annotated image from OmniParser
         """
         # Base system prompt
         is_mobile = platform in ("android", "ios")
@@ -41,18 +66,19 @@ def compose_do_messages(
                 "Your job: analyze the instruction and call the appropriate function to interact with the mobile UI.\n"
             )
             
-            # Add click guidance based on mode
-            if click_mode == "visual":
+            if element_source == "vision":
                 system_content += (
-                    "\nFOR CLICKING: Use click_visual_element(description) - describe the element visually.\n"
-                    "You will receive a screenshot. Analyze it and use visual descriptions.\n"
+                    "\nELEMENTS DETECTED VIA COMPUTER VISION (OmniParser):\n"
+                    "- tap_element(element_index): Click element by INDEX from numbered list\n"
+                    "- input_text(element_index, text): Type text into element by INDEX\n"
+                    "- The screenshot shows NUMBERED bounding boxes - use those numbers!\n"
                 )
-            else:  # xml (default)
+            else:
                 system_content += (
-                    "\nACTION SELECTION RULES:\n"
-                    "1. FOR TEXT INPUT: Use input_text(element_index, text) - select from numbered list\n"
-                    "2. FOR CLICKING: Use tap_element(index) - select from numbered list\n"
-                    "3. OTHER ACTIONS: scroll_down(), swipe_left/right/up(), long_press(index), hide_keyboard(), go_back()\n"
+                    "\nUSE LOCATOR TOOLS:\n"
+                    "1. FOR TEXT INPUT: input_text(element_index, text) - select from numbered list\n"
+                    "2. FOR CLICKING: tap_element(index) - select from numbered list\n"
+                    "3. OTHER: scroll_down(), swipe_left/right/up(), long_press(index), hide_keyboard(), go_back()\n"
                 )
             
             system_content += (
@@ -64,22 +90,21 @@ def compose_do_messages(
                 "Your job: analyze the instruction and call the appropriate function to interact with the web page.\n"
             )
             
-            # Add click guidance based on mode
-            if click_mode == "visual":
+            if element_source == "vision":
                 system_content += (
-                    "\nFOR INTERACTION: Use VISUAL tools:\n"
-                    "- click_visual_element(description): Click element by visual description\n"
-                    "- input_text_visual(description, text): Input text into element by visual description\n"
-                    "- hover_visual(description): Hover over element by visual description\n"
-                    "- double_click_visual(description): Double click element by visual description\n"
-                    "You will receive a screenshot. Analyze it and use visual descriptions.\n"
+                    "\nUSE VISUAL TOOLS:\n"
+                    "- click_visual_element(description): Click by visual description\n"
+                    "- input_text_visual(description, text): Input text by visual description\n"
+                    "- hover_visual(description): Hover by visual description\n"
+                    "- double_click_visual(description): Double click by visual description\n"
+                    "- Elements were detected using computer vision (OmniParser)\n"
                 )
-            else:  # xml (default)
+            else:
                 system_content += (
-                    "\nACTION SELECTION RULES:\n"
-                    "1. FOR TEXT INPUT: Use input_text(index, text) for <input> or <textarea> elements\n"
-                    "2. FOR CLICKING: Use click_element(index) for <button> or <a> elements\n"
-                    "3. FOR DROPDOWN: Use select_option(index, value) for <select> elements\n"
+                    "\nUSE LOCATOR TOOLS:\n"
+                    "1. FOR TEXT INPUT: input_text(index, text) for <input> or <textarea> elements\n"
+                    "2. FOR CLICKING: click_element(index) for <button> or <a> elements\n"
+                    "3. FOR DROPDOWN: select_option(index, value) for <select> elements\n"
                     "4. OTHER: scroll_down(), scroll_up(), press_key(), go_back(), hover(), double_click()\n"
                 )
 
@@ -90,33 +115,64 @@ def compose_do_messages(
                 "- <select> = dropdown (use select_option tool)\n"
             )
         
-        # Build user content based on input_mode
-        if click_mode == "visual":
-            user_content = f"Instruction: {instruction}\n\nAnalyze the screenshot and use visual description."
-        elif input_mode == "som" and screenshot_base64 and ui_elements:
-            # SoM mode: screenshot with numbered bounding boxes
-            from Agent.platforms.collectors.som_renderer import render_som
-            annotated_screenshot = render_som(screenshot_base64, ui_elements)
+        # Build user content based on llm_input_format
+        ui_label = "Mobile UI Elements" if is_mobile else "Web Elements"
+        
+        if llm_input_format == "som" and ui_elements:
+            source_info = "detected via computer vision" if element_source == "vision" else "from accessibility tree"
+            
+            legend_lines = []
+            for idx, elem in enumerate(ui_elements, start=1):
+                text = elem.get("text", "").replace("\n", " ").strip()[:40]
+                tag = elem.get("class_name", "")
+                short_tag = tag.split('.')[-1] if '.' in tag else tag
+                desc = text if text else (elem.get("aria_label") or elem.get("content_desc") or elem.get("placeholder") or "")
+                bbox = elem.get("bbox", {})
+                pos_info = ""
+                if bbox:
+                    y = bbox.get("y", 0)
+                    x = bbox.get("x", 0)
+                    pos = "top" if y < 400 else "mid" if y < 1200 else "bot"
+                    side = "L" if x < 300 else "C" if x < 700 else "R"
+                    pos_info = f" @{pos}-{side}"
+                legend_lines.append(f"[{idx}] {short_tag}: {desc}{pos_info}".strip())
+            legend_text = "\n".join(legend_lines)
             
-            ui_label = "Mobile UI Elements" if is_mobile else "Web Elements"
             text_content = (
                 f"Instruction: {instruction}\n\n"
-                f"The screenshot shows numbered boxes on {ui_label.lower()}.\n"
-                f"Select by NUMBER from the boxes shown in the image."
+                f"ANNOTATED SCREENSHOT: Each UI element has a GREEN BOX with its ID NUMBER in a small rectangle at the top-left.\n"
+                f"ELEMENT LIST ({source_info}):\n{legend_text}\n\n"
+                f"IMPORTANT: Select the element by its ID NUMBER that best matches the instruction."
             )
-            user_content = [
-                {"type": "text", "text": text_content},
-                {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{annotated_screenshot}"}}
-            ]
+            
+            # Use pre-annotated image from OmniParser if available (Visual + SoM)
+            if annotated_image_path:
+                with open(annotated_image_path, "rb") as img_file:
+                    annotated_base64 = base64.b64encode(img_file.read()).decode("utf-8")
+                self._save_annotated_image(annotated_base64, source="omniparser")
+                user_content = [
+                    {"type": "text", "text": text_content},
+                    {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{annotated_base64}"}}
+                ]
+            # Otherwise render SoM for DOM elements (DOM + SoM)
+            elif screenshot_base64:
+                from Agent.platforms.collectors.som_renderer import render_som
+                annotated_screenshot = render_som(screenshot_base64, ui_elements)
+                self._save_annotated_image(annotated_screenshot, source="dom")
+                user_content = [
+                    {"type": "text", "text": text_content},
+                    {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{annotated_screenshot}"}}
+                ]
+            else:
+                user_content = f"Instruction: {instruction}\n\nError: SoM mode requires screenshot"
         else:
-            # Text mode: numbered list
             if self.platform and ui_elements:
                 ui_text = self.platform.render_ui_for_prompt(ui_elements)
             else:
                 ui_text = "(no UI elements found)"
             
-            ui_label = "Mobile UI Elements" if is_mobile else "Web Elements"
-            user_content = f"Instruction: {instruction}\n\n{ui_label}:\n{ui_text}"
+            source_info = " (detected via OmniParser)" if element_source == "vision" else ""
+            user_content = f"Instruction: {instruction}\n\n{ui_label}{source_info}:\n{ui_text}"
         
         return [
             {"role": "system", "content": system_content},
@@ -126,18 +182,15 @@ def compose_do_messages(
     def get_do_tools(
         self, 
         category: str = "mobile",
-        click_mode: str = "xml"
+        element_source: str = "accessibility"
     ) -> List[Dict[str, Any]]:
         """Return tool definitions for DO actions from the registry.
         
-        Returns tool specs in standard format (works with OpenAI, Anthropic, Gemini, etc.)
-        
         Args:
-            category: Tool category to retrieve ('mobile', 'web', or None for all)
-            click_mode: 'xml' or 'visual' - filters tools based on capabilities
+            category: Tool category ('mobile')
+            element_source: 'accessibility' or 'vision'
         """
-        # Use registry's new filtering method
-        filtered_tools = self.registry.get_tools_for_mode(category, click_mode)
+        filtered_tools = self.registry.get_tools_for_source(category, element_source)
         return [tool.to_tool_spec() for tool in filtered_tools]
 
     def compose_visual_check_messages(
diff --git a/Agent/ai/llm/_anthropic.py b/Agent/ai/llm/_anthropic.py
index 016b5ef..ea52ce3 100644
--- a/Agent/ai/llm/_anthropic.py
+++ b/Agent/ai/llm/_anthropic.py
@@ -77,9 +77,9 @@ def create_chat_completion(
                 else:
                     api_params["tools"] = tools
 
-            if tool_choice:
-                from Agent.ai.llm._converters import convert_tool_choice_to_anthropic
-                api_params["tool_choice"] = convert_tool_choice_to_anthropic(tool_choice)
+                if tool_choice:
+                    from Agent.ai.llm._converters import convert_tool_choice_to_anthropic
+                    api_params["tool_choice"] = convert_tool_choice_to_anthropic(tool_choice)
 
             response = self.client.messages.create(**api_params)
 
diff --git a/Agent/ai/llm/_deepseek.py b/Agent/ai/llm/_deepseek.py
deleted file mode 100644
index bd6c150..0000000
--- a/Agent/ai/llm/_deepseek.py
+++ /dev/null
@@ -1,350 +0,0 @@
-from anthropic import Anthropic, APIError
-from typing import Optional, Dict, List, Union
-import os
-from robot.api import logger
-from Agent.ai.llm._baseclient import BaseLLMClient
-from Agent.config.model_config import ModelConfig
-from Agent.utilities._costtracker import CostTracker
-
-
-class DeepSeekClient(BaseLLMClient):
-    """
-    DeepSeek client using Anthropic API compatibility.
-    """
-
-    def __init__(
-        self, 
-        api_key: Optional[str] = None,
-        model: str = "deepseek-chat",
-        max_retries: int = 3,
-    ):
-        self.api_key: str = api_key
-        
-        if not self.api_key:
-            from Agent.config.config import Config
-            config = Config()
-            self.api_key = config.DEEPSEEK_API_KEY
-            logger.debug(f"API key loaded from config file")
-            
-        if not self.api_key:
-            raise ValueError("API key must be provided either as an argument or in the environment variables.")
-            
-        self.default_model = model
-        self.max_retries = max_retries
-        self.model_config = ModelConfig()
-        self.cost_tracker = CostTracker()
-        
-        # Initialize Anthropic client with DeepSeek's base URL
-        self.client = Anthropic(
-            api_key=self.api_key,
-            base_url="https://api.deepseek.com/anthropic",
-            max_retries=max_retries
-        )
-
-    def create_chat_completion(
-        self,
-        messages: List[Dict[str, str]],
-        model: Optional[str] = None,
-        max_tokens: int = 1400,
-        temperature: float = 1.0,
-        top_p: float = 1.0,
-        tools: Optional[List[Dict]] = None,
-        tool_choice: Optional[Union[str, Dict]] = None,
-        **kwargs
-    ):
-        """
-        Create a chat completion using DeepSeek's Anthropic-compatible API.
-        
-        Args:
-            messages: List of message dictionaries with 'role' and 'content'
-            model: Model to use (if None, uses default_model)
-            max_tokens: Maximum tokens to generate
-            temperature: Sampling temperature (0-2.0 for DeepSeek)
-            top_p: Nucleus sampling parameter (0-1)
-            tools: List of tools to use
-            tool_choice: Tool choice
-            **kwargs: Additional parameters to pass to the API
-            
-        Returns:
-            Anthropic Message object (from DeepSeek)
-        """
-        try:
-            self._validate_parameters(temperature, top_p)
-            
-            # DeepSeek follows Anthropic's format - separate system messages
-            system_message = None
-            user_messages = []
-            
-            for msg in messages:
-                if msg.get("role") == "system":
-                    system_message = msg.get("content")
-                else:
-                    transformed_content = self._transform_content(msg.get("content"))
-                    user_messages.append({
-                        "role": msg.get("role"),
-                        "content": transformed_content
-                    })
-            
-            # Prepare API call parameters
-            api_params = {
-                "model": model or self.default_model,
-                "messages": user_messages,
-                "max_tokens": max_tokens,
-                **kwargs
-            }
-            # Only add temperature or top_p, not both
-            if temperature != 1.0:
-                api_params["temperature"] = temperature
-            elif top_p != 1.0:
-                api_params["top_p"] = top_p
-            else:
-                # If both are default, use temperature
-                api_params["temperature"] = temperature
-            
-            if system_message:
-                api_params["system"] = system_message
-
-            if tools:
-                # Check format and convert if necessary
-                if len(tools) > 0 and isinstance(tools[0], dict) and "type" in tools[0] and tools[0]["type"] == "function":
-                    from Agent.ai.llm._converters import convert_to_anthropic_tools
-                    api_params["tools"] = convert_to_anthropic_tools(tools)
-                else:
-                    api_params["tools"] = tools
-            
-            if tool_choice:
-                from Agent.ai.llm._converters import convert_tool_choice_to_anthropic
-                api_params["tool_choice"] = convert_tool_choice_to_anthropic(tool_choice)
-            
-            response = self.client.messages.create(**api_params)
-            
-            # Log usage
-            logger.debug(
-                f"DeepSeek API call successful. Tokens used: {response.usage.input_tokens + response.usage.output_tokens}"
-            )
-            logger.debug(f"Response: {response}")
-            
-            return response
-            
-        except APIError as e:
-            logger.error(f"DeepSeek API Error: {str(e)}")
-            raise
-        except Exception as e:
-            logger.error(f"Unexpected error: {str(e)}")
-            raise
-
-    def _transform_content(self, content):
-        """
-        Transform content to DeepSeek's format (follows Anthropic format).
-        
-        Supports:
-        - String content (passed through)
-        - List content with text and images (if supported)
-        - OpenAI-style image_url format -> Anthropic's image format
-        - Native Anthropic image format (passed through)
-        
-        Args:
-            content: String or list of content items
-            
-        Returns:
-            Transformed content in Anthropic/DeepSeek format
-        """
-        # If content is a simple string, return as-is
-        if isinstance(content, str):
-            return content
-        
-        # If content is not a list, return as-is
-        if not isinstance(content, list):
-            return content
-        
-        # Transform list content
-        transformed = []
-        for item in content:
-            if not isinstance(item, dict):
-                continue
-                
-            item_type = item.get("type")
-            
-            # Handle text content
-            if item_type == "text":
-                transformed.append({
-                    "type": "text",
-                    "text": item.get("text", "")
-                })
-            
-            # Handle OpenAI-style image_url format
-            # Note: DeepSeek's docs show image support as "Not Supported"
-            # but we'll include the transformation for future compatibility
-            elif item_type == "image_url":
-                image_url_data = item.get("image_url", {})
-                if isinstance(image_url_data, dict) and "url" in image_url_data:
-                    url = image_url_data["url"]
-                    
-                    # Check if it's a base64 data URL
-                    if url.startswith("data:"):
-                        # Extract media type and base64 data
-                        try:
-                            header, data = url.split(",", 1)
-                            media_type = header.split(";")[0].split(":")[1]
-                            transformed.append({
-                                "type": "image",
-                                "source": {
-                                    "type": "base64",
-                                    "media_type": media_type,
-                                    "data": data
-                                }
-                            })
-                        except (ValueError, IndexError) as e:
-                            logger.error(f"Invalid base64 image URL format: {e}")
-                            continue
-                    else:
-                        # Regular URL
-                        transformed.append({
-                            "type": "image",
-                            "source": {
-                                "type": "url",
-                                "url": url
-                            }
-                        })
-            
-            # Handle native Anthropic image format
-            elif item_type == "image":
-                if "source" in item:
-                    transformed.append(item)
-                else:
-                    logger.warn("Image item missing 'source' field")
-            
-            # Pass through any other content types
-            else:
-                transformed.append(item)
-        
-        return transformed if transformed else content
-
-    def _validate_parameters(self, temperature: float, top_p: float):
-        """Validate API parameters. DeepSeek supports temperature 0-2.0."""
-        if not (0 <= temperature <= 2.0):
-            logger.error(f"Invalid temperature {temperature}. Must be between 0 and 2.0 for DeepSeek")
-            raise ValueError(f"Invalid temperature {temperature}. Must be between 0 and 2.0")
-        if not (0 <= top_p <= 1):
-            logger.error(f"Invalid top_p {top_p}. Must be between 0 and 1")
-            raise ValueError(f"Invalid top_p {top_p}. Must be between 0 and 1")
-    
-    def _calculate_cost(self, model: str, prompt_tokens: int, completion_tokens: int) -> Dict[str, float]:
-        """
-        Calculate the cost of API call based on token usage.
-        
-        Args:
-            model: Model name used for the API call
-            prompt_tokens: Number of input tokens
-            completion_tokens: Number of output tokens
-            
-        Returns:
-            Dictionary with input_cost, output_cost, and total_cost
-        """
-        pricing = self.model_config.get_model_pricing(model)
-        
-        if not pricing:
-            logger.warn(f"No pricing information found for model: {model}. Cost will be 0.")
-            return {
-                'input_cost': 0.0,
-                'output_cost': 0.0,
-                'total_cost': 0.0
-            }
-        
-        # Pricing is per 1M tokens according to llm_models.json metadata
-        input_cost = (prompt_tokens / 1_000_000) * pricing['input']
-        output_cost = (completion_tokens / 1_000_000) * pricing['output']
-        total_cost = input_cost + output_cost
-        
-        return {
-            'input_cost': input_cost,
-            'output_cost': output_cost,
-            'total_cost': total_cost
-        }
-
-    def format_response(
-        self, 
-        response,
-        include_tokens: bool = True,
-        include_reason: bool = False
-    ) -> Dict[str, Union[str, int]]:
-        """
-        Format DeepSeek response to a standardized dictionary.
-        
-        Args:
-            response: Anthropic Message object (from DeepSeek)
-            include_tokens: Whether to include token usage information
-            include_reason: Whether to include stop reason
-            
-        Returns:
-            Standardized response dictionary
-        """
-        if not response or not response.content:
-            logger.error(f"Invalid response or no content in the response")
-            return {}
-        
-        # Extract text content and tool calls (follows Anthropic's format)
-        content_text = ""
-        tool_calls = []
-        
-        for block in response.content:
-            if hasattr(block, 'text'):
-                content_text += block.text
-            elif hasattr(block, "type") and block.type == "tool_use":
-                # Anthropic/DeepSeek tool_use block
-                import json
-                tool_calls.append({
-                    "id": block.id,
-                    "type": "function",
-                    "function": {
-                        "name": block.name,
-                        "arguments": json.dumps(block.input)
-                    }
-                })
-        
-        result = {
-            "content": content_text,
-        }
-
-        if tool_calls:
-            result["tool_calls"] = tool_calls
-        
-        if include_tokens and response.usage:
-            logger.debug(f"Tokens used: input={response.usage.input_tokens}, output={response.usage.output_tokens}")
-            result.update({
-                "prompt_tokens": response.usage.input_tokens,
-                "completion_tokens": response.usage.output_tokens,
-                "total_tokens": response.usage.input_tokens + response.usage.output_tokens
-            })
-            
-            # Calculate and track cost
-            cost_data = self._calculate_cost(
-                model=response.model,
-                prompt_tokens=response.usage.input_tokens,
-                completion_tokens=response.usage.output_tokens
-            )
-            
-            result.update({
-                "input_cost": cost_data['input_cost'],
-                "output_cost": cost_data['output_cost'],
-                "total_cost": cost_data['total_cost']
-            })
-            
-            # Track cost in the cost tracker
-            self.cost_tracker.add_cost(
-                input_cost=cost_data['input_cost'],
-                output_cost=cost_data['output_cost'],
-                model=response.model
-            )
-            
-            logger.debug(
-                f"API call cost: ${cost_data['total_cost']:.6f} "
-                f"(input: ${cost_data['input_cost']:.6f}, output: ${cost_data['output_cost']:.6f})"
-            )
-            
-        if include_reason:
-            logger.debug(f"Stop reason: {response.stop_reason}")
-            result["finish_reason"] = response.stop_reason
-            
-        return result
-
diff --git a/Agent/ai/llm/_factory.py b/Agent/ai/llm/_factory.py
index 367fb95..3640107 100644
--- a/Agent/ai/llm/_factory.py
+++ b/Agent/ai/llm/_factory.py
@@ -1,17 +1,12 @@
 from typing import Optional, Callable, Dict
 from Agent.ai.llm._baseclient import BaseLLMClient
 from Agent.ai.llm._openaiclient import OpenAIClient
-from Agent.ai.llm._anthropic import AnthropicClient
-from Agent.ai.llm._gemini import GeminiClient
 from Agent.config.model_config import ModelConfig
 from Agent.config.config import Config
 
 
 class LLMClientFactory:
-    """
-    Factory class to create and return LLM client instances.
-    Supports multiple providers: OpenAI, Anthropic (Claude), and Google Gemini.
-    """
+    """Factory to create LLM client instances. Supports OpenAI, Anthropic, Gemini."""
 
     _model_config = ModelConfig()
     DEFAULT_MODELS = {
@@ -20,23 +15,38 @@ class LLMClientFactory:
         "gemini": _model_config.get_provider_default_model("gemini"),
     }
 
-    # Registry of provider factories for extension
     _registry: Dict[str, Callable[[Optional[str], Config], BaseLLMClient]] = {}
 
-    # Pre-register built-in providers
-    _registry.update({
-        "openai": lambda model, cfg: OpenAIClient(model=model or LLMClientFactory.DEFAULT_MODELS.get("openai"), api_key=cfg.OPENAI_API_KEY),
-        "anthropic": lambda model, cfg: AnthropicClient(model=model or LLMClientFactory.DEFAULT_MODELS.get("anthropic"), api_key=cfg.ANTHROPIC_API_KEY),
-        "claude": lambda model, cfg: AnthropicClient(model=model or LLMClientFactory.DEFAULT_MODELS.get("anthropic"), api_key=cfg.ANTHROPIC_API_KEY),
-        "gemini": lambda model, cfg: GeminiClient(model=model or LLMClientFactory.DEFAULT_MODELS.get("gemini")),
-        "google": lambda model, cfg: GeminiClient(model=model or LLMClientFactory.DEFAULT_MODELS.get("gemini")),
-    })
+    @staticmethod
+    def _create_anthropic_client(model: Optional[str], cfg: Config):
+        try:
+            from Agent.ai.llm._anthropic import AnthropicClient
+        except ImportError as e:
+            raise ImportError(
+                "anthropic package not installed. "
+                "Install with: pip install 'robotframework-agent[anthropic]'"
+            ) from e
+        return AnthropicClient(
+            model=model or LLMClientFactory.DEFAULT_MODELS.get("anthropic"),
+            api_key=cfg.ANTHROPIC_API_KEY
+        )
+    
+    @staticmethod
+    def _create_gemini_client(model: Optional[str], cfg: Config):
+        try:
+            from Agent.ai.llm._gemini import GeminiClient
+        except ImportError as e:
+            raise ImportError(
+                "google-generativeai package not installed. "
+                "Install with: pip install 'robotframework-agent[gemini]'"
+            ) from e
+        return GeminiClient(model=model or LLMClientFactory.DEFAULT_MODELS.get("gemini"))
 
     @staticmethod
     def register_client(name: str, factory: Callable[[Optional[str], Config], BaseLLMClient]) -> None:
-        """Register a custom provider factory callable.
-
-        The callable receives (model, config) and must return a BaseLLMClient instance.
+        """
+        Register custom provider: factory(model, config) -> BaseLLMClient
+        Example: factory.register_client("custom", lambda m, c: CustomClient(m))
         """
         LLMClientFactory._registry[name.lower()] = factory
 
@@ -45,10 +55,7 @@ def list_providers() -> Dict[str, str]:
         return {name: "registered" for name in LLMClientFactory._registry.keys()}
 
     @staticmethod
-    def create_client(
-        client_name: str = "openai",
-        model: Optional[str] = None,
-    ) -> BaseLLMClient:
+    def create_client(client_name: str = "openai", model: Optional[str] = None) -> BaseLLMClient:
         client_name_lower = client_name.lower()
         config = Config()
 
@@ -61,3 +68,13 @@ def create_client(
         return factory(model, config)
 
 
+LLMClientFactory._registry.update({
+    "openai": lambda model, cfg: OpenAIClient(
+        model=model or LLMClientFactory.DEFAULT_MODELS.get("openai"),
+        api_key=cfg.OPENAI_API_KEY
+    ),
+    "anthropic": lambda model, cfg: LLMClientFactory._create_anthropic_client(model, cfg),
+    "claude": lambda model, cfg: LLMClientFactory._create_anthropic_client(model, cfg),
+    "gemini": lambda model, cfg: LLMClientFactory._create_gemini_client(model, cfg),
+    "google": lambda model, cfg: LLMClientFactory._create_gemini_client(model, cfg),
+})
diff --git a/Agent/ai/llm/_gemini.py b/Agent/ai/llm/_gemini.py
index baf6ff2..8068924 100644
--- a/Agent/ai/llm/_gemini.py
+++ b/Agent/ai/llm/_gemini.py
@@ -73,10 +73,10 @@ def create_chat_completion(
                     request_params["tools"] = convert_to_gemini_tools(tools)
                 else:
                     request_params["tools"] = tools
-            
-            if tool_choice:
-                from Agent.ai.llm._converters import convert_tool_choice_to_gemini_config
-                request_params["tool_config"] = convert_tool_choice_to_gemini_config(tool_choice)
+                
+                if tool_choice:
+                    from Agent.ai.llm._converters import convert_tool_choice_to_gemini_config
+                    request_params["tool_config"] = convert_tool_choice_to_gemini_config(tool_choice)
 
             response = client.generate_content(**request_params)
 
diff --git a/Agent/ai/llm/_ollama.py b/Agent/ai/llm/_ollama.py
deleted file mode 100644
index b50c9ae..0000000
--- a/Agent/ai/llm/_ollama.py
+++ /dev/null
@@ -1,235 +0,0 @@
-from openai import OpenAI
-from openai.types.chat import ChatCompletion
-from typing import Optional, Dict, List, Union
-import os
-from robot.api import logger
-from Agent.ai.llm._baseclient import BaseLLMClient
-from Agent.config.model_config import ModelConfig
-from Agent.utilities._costtracker import CostTracker
-
-
-class OllamaClient(BaseLLMClient):
-    """
-    Ollama client for running local LLMs.
-    
-    Ollama provides a local server that's compatible with OpenAI's API format,
-    making it easy to run models like Llama, Mistral, CodeLlama, etc. locally.
-    
-    No API key required - just needs Ollama running locally.
-    Default endpoint: http://localhost:11434/v1
-    
-    Documentation: https://github.com/ollama/ollama/blob/main/docs/api.md
-    """
-    
-    def __init__(
-        self, 
-        model: str = "llama3.2",
-        base_url: str = "http://localhost:11434/v1",
-        max_retries: int = 3,
-    ):
-        self.default_model = model
-        self.base_url = base_url
-        self.max_retries = max_retries
-        self.model_config = ModelConfig()
-        self.cost_tracker = CostTracker()
-        
-        # Ollama doesn't require API key, but OpenAI SDK needs something
-        # Using "ollama" as a placeholder
-        self.client = OpenAI(
-            base_url=base_url,
-            api_key="ollama",  # Dummy key, not used by Ollama
-            max_retries=max_retries
-        )
-        
-        logger.debug(f"Ollama client initialized with base_url: {base_url}")
-
-    def create_chat_completion(
-        self,
-        messages: List[Dict[str, str]],
-        model: Optional[str] = None,
-        max_tokens: int = 1400,
-        temperature: float = 1.0,
-        top_p: float = 1.0,
-        tools: Optional[List[Dict]] = None,
-        tool_choice: Optional[Union[str, Dict]] = None,
-        **kwargs
-    ) -> Optional[ChatCompletion]:
-        """
-        Create a chat completion using local Ollama server.
-        
-        Args:
-            messages: List of message dictionaries with 'role' and 'content'
-            model: Model to use (if None, uses default_model)
-            max_tokens: Maximum tokens to generate
-            temperature: Sampling temperature (0-2)
-            top_p: Nucleus sampling parameter (0-1)
-            tools: List of tools to use
-            tool_choice: Tool choice
-            **kwargs: Additional parameters to pass to the API
-            
-        Returns:
-            OpenAI ChatCompletion object (Ollama is compatible)
-        """
-        try:
-            self._validate_parameters(temperature, top_p)
-            
-            params = {
-                "model": model or self.default_model,
-                "messages": messages,
-                "temperature": temperature,
-                "top_p": top_p,
-                "max_tokens": max_tokens,
-                **kwargs
-            }
-
-            if tools:
-                params["tools"] = tools
-            if tool_choice:
-                params["tool_choice"] = tool_choice
-
-            response = self.client.chat.completions.create(**params)
-            
-            # Log usage
-            logger.debug(
-                f"Ollama API call successful. Tokens used: {response.usage.total_tokens}",
-                True
-            )
-            logger.debug(f"Response: {response}")
-            
-            return response
-            
-        except Exception as e:
-            error_msg = str(e)
-            if "Connection" in error_msg or "refused" in error_msg:
-                logger.error(
-                    "Cannot connect to Ollama server. Is Ollama running? "
-                    "Start it with: ollama serve",
-                    True
-                )
-            else:
-                logger.error(f"Ollama API Error: {error_msg}")
-            raise
-
-    def _validate_parameters(self, temperature: float, top_p: float):
-        """Validate API parameters."""
-        if not (0 <= temperature <= 2):
-            logger.error(f"Invalid temperature {temperature}. Must be between 0 and 2")
-            raise ValueError(f"Invalid temperature {temperature}. Must be between 0 and 2")
-        if not (0 <= top_p <= 1):
-            logger.error(f"Invalid top_p {top_p}. Must be between 0 and 1")
-            raise ValueError(f"Invalid top_p {top_p}. Must be between 0 and 1")
-    
-    def _calculate_cost(self, model: str, prompt_tokens: int, completion_tokens: int) -> Dict[str, float]:
-        """
-        Calculate the cost of API call based on token usage.
-        Ollama is free (local), so costs are always 0, but we track for consistency.
-        
-        Args:
-            model: Model name used for the API call
-            prompt_tokens: Number of input tokens
-            completion_tokens: Number of output tokens
-            
-        Returns:
-            Dictionary with input_cost, output_cost, and total_cost (all 0 for Ollama)
-        """
-        pricing = self.model_config.get_model_pricing(model)
-        
-        if not pricing:
-            logger.debug(f"No pricing information found for model: {model}. Using 0 cost (local model).")
-            return {
-                'input_cost': 0.0,
-                'output_cost': 0.0,
-                'total_cost': 0.0
-            }
-        
-        # Pricing is per 1M tokens according to llm_models.json metadata
-        # For Ollama models, pricing should be 0.0, but we calculate anyway for consistency
-        input_cost = (prompt_tokens / 1_000_000) * pricing.get('input', 0.0)
-        output_cost = (completion_tokens / 1_000_000) * pricing.get('output', 0.0)
-        total_cost = input_cost + output_cost
-        
-        return {
-            'input_cost': input_cost,
-            'output_cost': output_cost,
-            'total_cost': total_cost
-        }
-
-    def format_response(
-        self, 
-        response: ChatCompletion,
-        include_tokens: bool = True,
-        include_reason: bool = False
-    ) -> Dict[str, Union[str, int]]:
-        """
-        Format Ollama response to a standardized dictionary.
-        
-        Args:
-            response: OpenAI ChatCompletion object (from Ollama)
-            include_tokens: Whether to include token usage information
-            include_reason: Whether to include stop reason
-            
-        Returns:
-            Standardized response dictionary
-        """
-        if not response or not response.choices:
-            logger.error(f"Invalid response or no choices in the response")
-            return {}
-            
-        result = {
-            "content": response.choices[0].message.content or "",
-        }
-
-        # Extract tool calls if present
-        if response.choices[0].message.tool_calls:
-            tool_calls = []
-            for tc in response.choices[0].message.tool_calls:
-                tool_calls.append({
-                    "id": tc.id,
-                    "type": tc.type,
-                    "function": {
-                        "name": tc.function.name,
-                        "arguments": tc.function.arguments
-                    }
-                })
-            result["tool_calls"] = tool_calls
-        
-        if include_tokens and response.usage:
-            logger.debug(f"Tokens used: {response.usage}")
-            result.update({
-                "prompt_tokens": response.usage.prompt_tokens,
-                "completion_tokens": response.usage.completion_tokens,
-                "total_tokens": response.usage.total_tokens
-            })
-            
-            # Calculate and track cost (Ollama is free, but we track for consistency)
-            cost_data = self._calculate_cost(
-                model=response.model,
-                prompt_tokens=response.usage.prompt_tokens,
-                completion_tokens=response.usage.completion_tokens
-            )
-            
-            result.update({
-                "input_cost": cost_data['input_cost'],
-                "output_cost": cost_data['output_cost'],
-                "total_cost": cost_data['total_cost']
-            })
-            
-            # Track cost in the cost tracker (will be 0 for local Ollama)
-            self.cost_tracker.add_cost(
-                input_cost=cost_data['input_cost'],
-                output_cost=cost_data['output_cost'],
-                model=response.model
-            )
-            
-            logger.debug(
-                f"API call cost: ${cost_data['total_cost']:.6f} "
-                f"(input: ${cost_data['input_cost']:.6f}, output: ${cost_data['output_cost']:.6f}) "
-                f"[Local Ollama - Free]"
-            )
-            
-        if include_reason:
-            logger.debug(f"Finish reason: {response.choices[0].finish_reason}")
-            result["finish_reason"] = response.choices[0].finish_reason
-            
-        return result
-
diff --git a/Agent/ai/llm/_openaiclient.py b/Agent/ai/llm/_openaiclient.py
index 7106c02..5e016d7 100644
--- a/Agent/ai/llm/_openaiclient.py
+++ b/Agent/ai/llm/_openaiclient.py
@@ -54,8 +54,8 @@ def create_chat_completion(
 
             if tools:
                 params["tools"] = tools
-            if tool_choice:
-                params["tool_choice"] = tool_choice
+                if tool_choice:
+                    params["tool_choice"] = tool_choice
 
             response = self.client.chat.completions.create(**params)
             logger.debug(f"OpenAI API call successful. Tokens used: {response.usage.total_tokens}")
@@ -135,7 +135,7 @@ def format_response(
             result["tool_calls"] = tool_calls
 
         if include_tokens:
-            logger.info(f"Tokens used: {response.usage}")
+            logger.debug(f"Tokens used: {response.usage}")
             result.update(
                 {
                     "prompt_tokens": response.usage.prompt_tokens,
diff --git a/Agent/ai/llm/facade.py b/Agent/ai/llm/facade.py
index 97b3ee5..dac48a3 100644
--- a/Agent/ai/llm/facade.py
+++ b/Agent/ai/llm/facade.py
@@ -1,11 +1,25 @@
 from typing import Any, Dict, List, Optional
 import json
+import copy
 
 from robot.api import logger
 from Agent.utilities._jsonutils import extract_json_safely
 from Agent.ai.llm._factory import LLMClientFactory
 from Agent.ai.llm._openaiclient import OpenAIClient
-from Agent.ai.llm._ollama import OllamaClient
+
+
+def _sanitize_for_log(messages: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
+    """Remove base64 images from messages for readable logging."""
+    sanitized = copy.deepcopy(messages)
+    for msg in sanitized:
+        content = msg.get("content")
+        if isinstance(content, list):
+            for item in content:
+                if isinstance(item, dict) and item.get("type") == "image_url":
+                    url = item.get("image_url", {}).get("url", "")
+                    if url.startswith("data:image"):
+                        item["image_url"]["url"] = "[BASE64_IMAGE]"
+    return sanitized
 
 
 class UnifiedLLMFacade:
@@ -43,10 +57,10 @@ def send_ai_request_and_return_response(
     ) -> Dict[str, Any]:
         """Sends a request to the AI model and returns a parsed JSON response."""
         logger.debug("🚀 Sending request to AI model...")
-        logger.debug(f"📝 Prompt: {json.dumps(messages, indent=2)}")
+        logger.debug(f"📝 Prompt: {json.dumps(_sanitize_for_log(messages), indent=2)}")
         
-        # Only OpenAI and Ollama support response_format parameter
-        if isinstance(self._client, (OpenAIClient, OllamaClient)):
+        # Only OpenAI supports response_format parameter
+        if isinstance(self._client, OpenAIClient):
             kwargs["response_format"] = {"type": "json_object"}
         
         response = self._client.create_chat_completion(
@@ -72,7 +86,7 @@ def send_ai_request_with_tools(
     ) -> Dict[str, Any]:
         """Sends a request to the AI model with tool calling enabled."""
         logger.debug("🚀 Sending request to AI model with tools...")
-        logger.debug(f"📝 Prompt: {json.dumps(messages, indent=2)}")
+        logger.debug(f"📝 Prompt: {json.dumps(_sanitize_for_log(messages), indent=2)}")
         logger.debug(f"   Tools: {[t['function']['name'] for t in tools]}")
         
         response = self._client.create_chat_completion(
diff --git a/Agent/ai/prompts/renderer.py b/Agent/ai/prompts/renderer.py
index ffd5a62..5c8fc7c 100644
--- a/Agent/ai/prompts/renderer.py
+++ b/Agent/ai/prompts/renderer.py
@@ -66,6 +66,10 @@ def _render_android_element(self, index: int, el: Dict[str, Any]) -> str:
         """Render a single Android element."""
         parts = []
         
+        class_name = el.get('class_name', 'unknown')
+        short_class = class_name.split('.')[-1] if '.' in class_name else class_name
+        parts.append(f"[{short_class}]")
+        
         if el.get("text"):
             parts.append(f"text='{el['text']}'")
         
@@ -74,18 +78,28 @@ def _render_android_element(self, index: int, el: Dict[str, Any]) -> str:
         
         content_desc = el.get("accessibility_label", '') or el.get("content_desc", '')
         if content_desc:
-            parts.append(f"content-desc='{content_desc}'")
-        
-        if parts:
-            return f"{index}. {' | '.join(parts)}"
+            parts.append(f"desc='{content_desc}'")
+        
+        bbox = el.get("bbox", {})
+        if bbox:
+            y = bbox.get("y", 0)
+            x = bbox.get("x", 0)
+            w = bbox.get("width", 0)
+            h = bbox.get("height", 0)
+            pos = "top" if y < 400 else "middle" if y < 1200 else "bottom"
+            side = "left" if x < 300 else "center" if x < 700 else "right"
+            parts.append(f"pos={pos}-{side} size={w}x{h}")
         
-        class_name = el.get('class_name', 'unknown')
-        return f"{index}. {class_name}"
+        return f"{index}. {' | '.join(parts)}"
     
     def _render_ios_element(self, index: int, el: Dict[str, Any]) -> str:
         """Render a single iOS element."""
         parts = []
         
+        class_name = el.get('class_name', 'unknown')
+        short_class = class_name.replace('XCUIElementType', '') if 'XCUIElementType' in class_name else class_name
+        parts.append(f"[{short_class}]")
+        
         if el.get("text"):
             parts.append(f"text='{el['text']}'")
         
@@ -96,8 +110,14 @@ def _render_ios_element(self, index: int, el: Dict[str, Any]) -> str:
         if label:
             parts.append(f"label='{label}'")
         
-        if parts:
-            return f"{index}. {' | '.join(parts)}"
+        bbox = el.get("bbox", {})
+        if bbox:
+            y = bbox.get("y", 0)
+            x = bbox.get("x", 0)
+            w = bbox.get("width", 0)
+            h = bbox.get("height", 0)
+            pos = "top" if y < 400 else "middle" if y < 1200 else "bottom"
+            side = "left" if x < 300 else "center" if x < 700 else "right"
+            parts.append(f"pos={pos}-{side} size={w}x{h}")
         
-        class_name = el.get('class_name', 'unknown')
-        return f"{index}. {class_name}"
+        return f"{index}. {' | '.join(parts)}"
diff --git a/Agent/ai/reademe.me b/Agent/ai/reademe.me
deleted file mode 100644
index 63b7198..0000000
--- a/Agent/ai/reademe.me
+++ /dev/null
@@ -1,5 +0,0 @@
-preferrably everything linked to ai heere 
-and preferrably one file interface for all dealign with ai ( currently _agentengine )
-
-always suggestions and other vision are welcome ( this is just a 
-second more clean iteration)
\ No newline at end of file
diff --git a/Agent/ai/vlm/_client.py b/Agent/ai/vlm/_client.py
index d64286e..4b634cc 100644
--- a/Agent/ai/vlm/_client.py
+++ b/Agent/ai/vlm/_client.py
@@ -36,13 +36,11 @@ def __init__(
         self.api_name = api_name or Config.OMNIPARSER_API_NAME
         self.hf_token = hf_token or Config.get_huggingface_token() or None
 
-        client_kwargs: Dict[str, Any] = {}
-        if self.hf_token:
-            # gradio_client 2.0+ uses 'token' instead of 'hf_token'
-            client_kwargs["token"] = self.hf_token
+        if not self.hf_token:
+            raise ValueError("HUGGINGFACE_API_KEY must be provided in environment variables.")
 
         try:
-            self._client = Client(self.space_id, **client_kwargs)
+            self._client = Client(self.space_id, token=self.hf_token)
             logger.debug(f"Initialized OmniParser client for space '{self.space_id}'")
         except Exception as exc:  # pragma: no cover - network failure
             raise OmniParserError(f"Failed to initialise OmniParser client: {exc}") from exc
diff --git a/Agent/ai/vlm/_parser.py b/Agent/ai/vlm/_parser.py
index 01b9f34..3087d39 100644
--- a/Agent/ai/vlm/_parser.py
+++ b/Agent/ai/vlm/_parser.py
@@ -162,3 +162,19 @@ def to_icon(self) -> Icon:
             interactivity=self.interactivity,
             content=self.content,
         )
+
+
+if __name__ == "__main__":
+    parser = OmniParserResultProcessor(
+        response_text="""
+        icon 1: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': True, 'content': 'YouTube'}
+        icon 2: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': False, 'content': 'YouTube'}
+        icon 3: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': True, 'content': 'YouTube'}
+        icon 4: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': False, 'content': 'YouTube'}
+        icon 5: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': False, 'content': 'YouTube'}
+        icon 6: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': False, 'content': 'YouTube'}
+        icon 7: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': False, 'content': 'YouTube'}
+        icon 8: {'type': 'icon', 'bbox': [0.419, 0.17, 0.574, 0.266], 'interactivity': False, 'content': 'YouTube'}
+        """
+    )
+    print(parser.get_parsed_ui_elements(element_type="interactive"))
\ No newline at end of file
diff --git a/Agent/ai/vlm/_selector.py b/Agent/ai/vlm/_selector.py
index 9383775..e938d57 100644
--- a/Agent/ai/vlm/_selector.py
+++ b/Agent/ai/vlm/_selector.py
@@ -172,4 +172,20 @@ def _parse_response(
             "element_data": elements_data[element_key],
             "confidence": response.get("confidence", "unknown"),
             "reason": response.get("reason", ""),
-        }
\ No newline at end of file
+        }
+
+
+if __name__ == "__main__":
+    selector = OmniParserElementSelector()
+    result = selector.select_element(
+        elements_data={
+            "icon3": {"type": "icon", "bbox": [0.419, 0.17, 0.574, 0.266], "interactivity": True, "content": "st"},
+            "icon4": {"type": "icon", "bbox": [0.419, 0.17, 0.574, 0.266], "interactivity": False, "content": "st"},
+            "icon5": {"type": "icon", "bbox": [0.419, 0.17, 0.574, 0.266], "interactivity": False, "content": "st"},
+            "icon6": {"type": "icon", "bbox": [0.419, 0.17, 0.574, 0.266], "interactivity": False, "content": "st"},
+            "icon7": {"type": "icon", "bbox": [0.419, 0.17, 0.574, 0.266], "interactivity": False, "content": "st"},
+            "icon8": {"type": "icon", "bbox": [0.419, 0.17, 0.574, 0.266], "interactivity": False, "content": "YouTube"},
+        },
+        element_description="YouTube icon",
+    )
+    print(result)
\ No newline at end of file
diff --git a/Agent/ai/vlm/interface.py b/Agent/ai/vlm/interface.py
index bc561af..54e378e 100644
--- a/Agent/ai/vlm/interface.py
+++ b/Agent/ai/vlm/interface.py
@@ -127,6 +127,79 @@ def find_element(
         
         return result
 
+    def detect_all_elements(
+        self,
+        image_base64: str,
+        element_type: str = "interactive",
+        box_threshold: Optional[float] = None,
+        iou_threshold: Optional[float] = None,
+        use_paddleocr: Optional[bool] = None,
+        imgsz: Optional[int] = None,
+    ) -> List[Dict[str, Any]]:
+        """
+        Detect all elements using OmniParser.
+        
+        Args:
+            image_base64: Base64 encoded image
+            element_type: Type filter ('interactive', 'text', 'icon', 'all')
+            
+        Returns:
+            List of elements in standard format compatible with DOM elements
+            Example: [{'text': 'Login', 'bbox': {'x': 10, 'y': 20, 'width': 100, 'height': 50}, 'source': 'omniparser'}, ...]
+        """
+        logger.debug(f"🔍 Detecting all elements with OmniParser (type={element_type})")
+        
+        image_temp_path, parsed_text = self.client.parse_image(
+            image_base64=image_base64,
+            box_threshold=box_threshold,
+            iou_threshold=iou_threshold,
+            use_paddleocr=use_paddleocr,
+            imgsz=imgsz,
+        )
+        
+        if not parsed_text:
+            logger.debug("❌ OmniParser detected no elements")
+            return []
+        
+        processor = OmniParserResultProcessor(
+            response_text=parsed_text,
+            image_temp_path=image_temp_path,
+        )
+        elements_data = processor.get_parsed_ui_elements(element_type=element_type)
+        
+        if not elements_data:
+            logger.debug(f"❌ No elements of type '{element_type}' found")
+            return []
+        
+        from PIL import Image
+        with Image.open(image_temp_path) as img:
+            width, height = img.size
+        
+        result = []
+        for key, data in elements_data.items():
+            bbox_norm = data.get("bbox", [0, 0, 0, 0])
+            x1 = int(bbox_norm[0] * width)
+            y1 = int(bbox_norm[1] * height)
+            x2 = int(bbox_norm[2] * width)
+            y2 = int(bbox_norm[3] * height)
+            
+            element = {
+                "text": data.get("content", ""),
+                "class_name": data.get("type", "unknown"),
+                "bbox": {
+                    "x": x1,
+                    "y": y1,
+                    "width": x2 - x1,
+                    "height": y2 - y1
+                },
+                "source": "omniparser",
+                "interactivity": data.get("interactivity", "unknown")
+            }
+            result.append(element)
+        
+        logger.debug(f"✅ Detected {len(result)} visual elements")
+        return result
+
     @staticmethod
     def get_element_center_coordinates(
         element_result: Dict[str, Any]
@@ -216,4 +289,13 @@ def bbox_to_pixels_from_image(
             bbox_normalized=bbox_normalized,
             image_width=width,
             image_height=height,
-        )
\ No newline at end of file
+        )
+
+
+if __name__ == "__main__":
+    orchestrator = OmniParserOrchestrator()
+    result = orchestrator.find_element(
+        element_description="YouTube icon",
+        image_path="tests/_data/mobilescreenshots/screenshot-Google Pixel 5-11.0.png"
+    )
+    print(result)
\ No newline at end of file
diff --git a/Agent/config/__init__.py b/Agent/config/__init__.py
index 305241c..cfe855a 100644
--- a/Agent/config/__init__.py
+++ b/Agent/config/__init__.py
@@ -1,7 +1,4 @@
-"""
-Configuration module for AiHelper.
-Provides access to environment configurations and model configurations.
-"""
+"""Configuration for Robot Framework Agent."""
 from .config import Config
 from .model_config import ModelConfig
 
diff --git a/Agent/config/config.py b/Agent/config/config.py
index 9d42da7..4cf59e3 100644
--- a/Agent/config/config.py
+++ b/Agent/config/config.py
@@ -4,22 +4,22 @@
 
 load_dotenv()
 
+_model_config = ModelConfig()
+
 
 class Config:
     """
-    Configuration class for AI Helper.
-    Loads API keys from environment variables (.env file).
-    Default models are loaded from llm_models.json (single source of truth).
+    Configuration for Robot Framework Agent.
+    Loads settings from environment variables (.env file).
     """
-    
+
     # LLM Provider Settings
     DEFAULT_LLM_CLIENT = os.getenv("DEFAULT_LLM_CLIENT", "openai")
-    
-    # API Keys for LLM Providers
+
+    # API Keys
     OPENAI_API_KEY = os.getenv("OPENAI_API_KEY", "")
     ANTHROPIC_API_KEY = os.getenv("ANTHROPIC_API_KEY", "")
     GEMINI_API_KEY = os.getenv("GEMINI_API_KEY", "")
-    DEEPSEEK_API_KEY = os.getenv("DEEPSEEK_API_KEY", "")
     HUGGINGFACE_API_KEY = os.getenv("HUGGINGFACE_API_KEY", "")
 
     # OmniParser (Hugging Face space)
@@ -28,27 +28,17 @@ class Config:
     OMNIPARSER_USE_PADDLE_OCR = os.getenv("OMNIPARSER_USE_PADDLE_OCR", "true").lower() == "true"
     OMNIPARSER_DEFAULT_BOX_THRESHOLD = float(os.getenv("OMNIPARSER_BOX_THRESHOLD", "0.25"))
     OMNIPARSER_DEFAULT_IOU_THRESHOLD = float(os.getenv("OMNIPARSER_IOU_THRESHOLD", "0.1"))
-    OMNIPARSER_DEFAULT_IMAGE_SIZE = int(float(os.getenv("OMNIPARSER_IMAGE_SIZE", "640")))
-    
-    # Default Models per Provider
-    _model_config = ModelConfig()
+    OMNIPARSER_DEFAULT_IMAGE_SIZE = int(os.getenv("OMNIPARSER_IMAGE_SIZE", "640"))
+
+    # Default Models
     DEFAULT_OPENAI_MODEL = os.getenv("DEFAULT_OPENAI_MODEL") or _model_config.get_provider_default_model("openai")
     DEFAULT_ANTHROPIC_MODEL = os.getenv("DEFAULT_ANTHROPIC_MODEL") or _model_config.get_provider_default_model("anthropic")
     DEFAULT_GEMINI_MODEL = os.getenv("DEFAULT_GEMINI_MODEL") or _model_config.get_provider_default_model("gemini")
-    DEFAULT_DEEPSEEK_MODEL = os.getenv("DEFAULT_DEEPSEEK_MODEL") or _model_config.get_provider_default_model("deepseek")
-    DEFAULT_OLLAMA_MODEL = os.getenv("DEFAULT_OLLAMA_MODEL") or _model_config.get_provider_default_model("ollama")
-    
-    # Ollama Configuration (local server)
-    OLLAMA_BASE_URL = os.getenv("OLLAMA_BASE_URL", "http://localhost:11434/v1")
-    
-    # Image Upload Provider API Keys
+
+    # Image Upload
     IMGBB_API_KEY = os.getenv("IMGBB_API_KEY", "")
     FREEIMAGEHOST_API_KEY = os.getenv("FREEIMAGEHOST_API_KEY", "")
 
-    # DEFAULT_MAX_TOKENS = 1400
-    # DEFAULT_TEMPERATURE = 1.0
-    # DEFAULT_TOP_P = 1.0
-
     @classmethod
     def get_huggingface_token(cls) -> str:
         return cls.HUGGINGFACE_API_KEY
@@ -61,10 +51,9 @@ def get_omniparser_params(
         use_paddleocr: bool | None = None,
         imgsz: int | None = None,
     ) -> dict[str, float | bool | int]:
-        params: dict[str, float | bool | int] = {
+        return {
             "box_threshold": box_threshold if box_threshold is not None else cls.OMNIPARSER_DEFAULT_BOX_THRESHOLD,
             "iou_threshold": iou_threshold if iou_threshold is not None else cls.OMNIPARSER_DEFAULT_IOU_THRESHOLD,
             "use_paddleocr": use_paddleocr if use_paddleocr is not None else cls.OMNIPARSER_USE_PADDLE_OCR,
             "imgsz": imgsz if imgsz is not None else cls.OMNIPARSER_DEFAULT_IMAGE_SIZE,
-        }
-        return params
\ No newline at end of file
+        }
\ No newline at end of file
diff --git a/Agent/config/llm_models.json b/Agent/config/llm_models.json
index f7eb31c..5522b0e 100644
--- a/Agent/config/llm_models.json
+++ b/Agent/config/llm_models.json
@@ -11,14 +11,6 @@
     "gemini": {
       "name": "Google Gemini",
       "default_model": "gemini-2.0-flash"
-    },
-    "deepseek": {
-      "name": "DeepSeek",
-      "default_model": "deepseek-chat"
-    },
-    "ollama": {
-      "name": "Ollama (Local)",
-      "default_model": "llama3.2"
     }
   },
   "models": {
@@ -195,15 +187,6 @@
       },
       "max_context_tokens": 200000
     },
-    "deepseek-chat": {
-      "provider": "deepseek",
-      "display_name": "Deepseek Chat",
-      "pricing": {
-        "input": 0.014,
-        "output": 0.028
-      },
-      "max_context_tokens": 65536
-    },
     "claude-3-5-haiku": {
       "provider": "gemini",
       "display_name": "Claude 3 5",
@@ -1905,96 +1888,6 @@
         "output": 0.0
       },
       "max_context_tokens": 32768
-    },
-    "llama3.2": {
-      "provider": "ollama",
-      "display_name": "Llama 3.2 (3B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 131072,
-      "notes": "Local model, free to use. Fast and efficient."
-    },
-    "llama3.2:1b": {
-      "provider": "ollama",
-      "display_name": "Llama 3.2 (1B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 131072,
-      "notes": "Local model, very lightweight and fast"
-    },
-    "llama3.1": {
-      "provider": "ollama",
-      "display_name": "Llama 3.1 (8B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 131072,
-      "notes": "Local model, better reasoning capabilities"
-    },
-    "llama3.3:70b": {
-      "provider": "ollama",
-      "display_name": "Llama 3.3 (70B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 131072,
-      "notes": "Local model, most powerful. Requires ~40GB RAM"
-    },
-    "mistral": {
-      "provider": "ollama",
-      "display_name": "Mistral (7B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 32768,
-      "notes": "Local model, balanced performance"
-    },
-    "codellama": {
-      "provider": "ollama",
-      "display_name": "Code Llama (7B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 16384,
-      "notes": "Local model, optimized for code generation"
-    },
-    "phi3": {
-      "provider": "ollama",
-      "display_name": "Phi-3 (3.8B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 131072,
-      "notes": "Local model, Microsoft's efficient model"
-    },
-    "gemma2": {
-      "provider": "ollama",
-      "display_name": "Gemma 2 (9B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 8192,
-      "notes": "Local model, Google's open model"
-    },
-    "deepseek-coder": {
-      "provider": "ollama",
-      "display_name": "DeepSeek Coder (6.7B)",
-      "pricing": {
-        "input": 0.0,
-        "output": 0.0
-      },
-      "max_context_tokens": 16384,
-      "notes": "Local model, specialized for coding tasks"
     }
   },
   "metadata": {
@@ -2004,8 +1897,7 @@
     "references": {
       "gemini": "https://ai.google.dev/gemini-api/docs/pricing",
       "anthropic": "https://docs.anthropic.com/en/docs/about-claude/models",
-      "openai": "https://openai.com/api/pricing/",
-      "deepseek": "https://www.deepseek.com/pricing"
+      "openai": "https://openai.com/api/pricing/"
     }
   }
 }
\ No newline at end of file
diff --git a/Agent/core/__init__.py b/Agent/core/__init__.py
index 2e2b041..9d32708 100644
--- a/Agent/core/__init__.py
+++ b/Agent/core/__init__.py
@@ -1,25 +1,7 @@
-from Agent.core.interfaces import (
-    PlatformProtocol,
-    CollectorProtocol,
-    LocatorProtocol,
-    ExecutorProtocol,
-)
-from Agent.core.models import (
-    BoundingBox,
-    WebElement,
-    AndroidElement,
-    IOSElement,
-)
+from Agent.core.interfaces import ExecutorProtocol
 from Agent.core.keyword_runner import KeywordRunner
 
 __all__ = [
-    "PlatformProtocol",
-    "CollectorProtocol",
-    "LocatorProtocol",
     "ExecutorProtocol",
     "KeywordRunner",
-    "BoundingBox",
-    "WebElement",
-    "AndroidElement",
-    "IOSElement",
 ]
diff --git a/Agent/core/interfaces.py b/Agent/core/interfaces.py
index 19c93ff..8d8f22a 100644
--- a/Agent/core/interfaces.py
+++ b/Agent/core/interfaces.py
@@ -1,47 +1,4 @@
-from typing import Any, Dict, List, Protocol, runtime_checkable
-#TODO: review and plug
-
-@runtime_checkable
-class PlatformProtocol(Protocol):
-    """Protocol for platform connectors (web/mobile)."""
-    
-    def get_platform(self) -> str:
-        """Return platform name ('web', 'android', 'ios')."""
-        ...
-    
-    def get_screenshot_base64(self) -> str:
-        """Capture screenshot as base64 string."""
-        ...
-    
-    def collect_ui_candidates(self, max_items: int = 500) -> List[Dict[str, Any]]:
-        """Collect interactive UI elements."""
-        ...
-    
-    def embed_image_to_log(self, base64_screenshot: str, width: int = 400) -> None:
-        """Embed image into Robot Framework log."""
-        ...
-
-
-@runtime_checkable
-class CollectorProtocol(Protocol):
-    """Protocol for UI element collectors."""
-    
-    def collect_elements(self, max_items: int = 500) -> List[Dict[str, Any]]:
-        """Collect interactive UI elements from current screen/page."""
-        ...
-    
-    def get_name(self) -> str:
-        """Return collector strategy name."""
-        ...
-
-
-@runtime_checkable
-class LocatorProtocol(Protocol):
-    """Protocol for locator builders."""
-    
-    def build(self, element: Dict[str, Any]) -> str:
-        """Build locator string from element attributes."""
-        ...
+from typing import Any, Dict, Protocol, runtime_checkable
 
 
 @runtime_checkable
@@ -49,10 +6,7 @@ class ExecutorProtocol(Protocol):
     """Protocol for action executors."""
     
     def run_keyword(self, keyword_name: str, *args: Any) -> Any:
-        """Execute a Robot Framework keyword."""
         ...
     
     def build_locator(self, element: Dict[str, Any]) -> str:
-        """Build locator from element attributes."""
         ...
-
diff --git a/Agent/core/keyword_runner.py b/Agent/core/keyword_runner.py
index e571d93..1643dd2 100644
--- a/Agent/core/keyword_runner.py
+++ b/Agent/core/keyword_runner.py
@@ -7,6 +7,13 @@ class KeywordRunner:
     
     def __init__(self, platform) -> None:
         self.platform = platform
+        self._appium = None
+    
+    @property
+    def appium(self):
+        if self._appium is None:
+            self._appium = BuiltIn().get_library_instance('AppiumLibrary')
+        return self._appium
     
     def run_keyword(self, keyword_name: str, *args: Any) -> Any:
         return BuiltIn().run_keyword(keyword_name, *args)
diff --git a/Agent/core/models.py b/Agent/core/models.py
deleted file mode 100644
index 779eae1..0000000
--- a/Agent/core/models.py
+++ /dev/null
@@ -1,158 +0,0 @@
-# TODO: clean, validate, and plug
-
-from dataclasses import dataclass, field
-from typing import Dict, Any
-
-
-@dataclass
-class BoundingBox:
-    x: int = 0
-    y: int = 0
-    width: int = 0
-    height: int = 0
-    
-    def is_valid(self) -> bool:
-        return self.width > 0 and self.height > 0
-    
-    def to_dict(self) -> Dict[str, int]:
-        return {"x": self.x, "y": self.y, "width": self.width, "height": self.height}
-    
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "BoundingBox":
-        if not data:
-            return cls()
-        return cls(
-            x=int(data.get("x", 0)),
-            y=int(data.get("y", 0)),
-            width=int(data.get("width", 0)),
-            height=int(data.get("height", 0)),
-        )
-
-
-@dataclass
-class WebElement:
-    """Web element (Playwright)."""
-    text: str = ""
-    element_id: str = ""
-    tag: str = ""
-    aria_label: str = ""
-    placeholder: str = ""
-    css_class: str = ""
-    href: str = ""
-    role: str = ""
-    name: str = ""
-    element_type: str = ""
-    clickable: bool = True
-    enabled: bool = True
-    bbox: BoundingBox = field(default_factory=BoundingBox)
-    
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "text": self.text,
-            "resource_id": self.element_id,
-            "class_name": self.tag,
-            "aria_label": self.aria_label,
-            "placeholder": self.placeholder,
-            "css_class": self.css_class,
-            "href": self.href,
-            "role": self.role,
-            "name": self.name,
-            "type": self.element_type,
-            "clickable": self.clickable,
-            "enabled": self.enabled,
-            "bbox": self.bbox.to_dict(),
-        }
-    
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "WebElement":
-        bbox = BoundingBox.from_dict(data.get("bbox", {}))
-        return cls(
-            text=data.get("text", ""),
-            element_id=data.get("resource_id", "") or data.get("element_id", ""),
-            tag=data.get("class_name", "") or data.get("tag", ""),
-            aria_label=data.get("aria_label", ""),
-            placeholder=data.get("placeholder", ""),
-            css_class=data.get("css_class", ""),
-            href=data.get("href", ""),
-            role=data.get("role", ""),
-            name=data.get("name", ""),
-            element_type=data.get("type", "") or data.get("element_type", ""),
-            clickable=data.get("clickable", True),
-            enabled=data.get("enabled", True),
-            bbox=bbox,
-        )
-
-
-@dataclass
-class AndroidElement:
-    """Android element (Appium)."""
-    text: str = ""
-    resource_id: str = ""
-    class_name: str = ""
-    content_desc: str = ""
-    clickable: bool = True
-    enabled: bool = True
-    bbox: BoundingBox = field(default_factory=BoundingBox)
-    
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "text": self.text,
-            "resource_id": self.resource_id,
-            "class_name": self.class_name,
-            "content_desc": self.content_desc,
-            "accessibility_label": self.content_desc,
-            "clickable": self.clickable,
-            "enabled": self.enabled,
-            "bbox": self.bbox.to_dict(),
-        }
-    
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "AndroidElement":
-        bbox = BoundingBox.from_dict(data.get("bbox", {}))
-        return cls(
-            text=data.get("text", ""),
-            resource_id=data.get("resource_id", ""),
-            class_name=data.get("class_name", ""),
-            content_desc=data.get("content_desc", "") or data.get("accessibility_label", ""),
-            clickable=data.get("clickable", True),
-            enabled=data.get("enabled", True),
-            bbox=bbox,
-        )
-
-
-@dataclass
-class IOSElement:
-    """iOS element (Appium)."""
-    text: str = ""
-    name: str = ""
-    label: str = ""
-    element_type: str = ""
-    clickable: bool = True
-    enabled: bool = True
-    bbox: BoundingBox = field(default_factory=BoundingBox)
-    
-    def to_dict(self) -> Dict[str, Any]:
-        return {
-            "text": self.text,
-            "resource_id": self.name,
-            "name": self.name,
-            "label": self.label,
-            "accessibility_label": self.label,
-            "class_name": self.element_type,
-            "clickable": self.clickable,
-            "enabled": self.enabled,
-            "bbox": self.bbox.to_dict(),
-        }
-    
-    @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "IOSElement":
-        bbox = BoundingBox.from_dict(data.get("bbox", {}))
-        return cls(
-            text=data.get("text", ""),
-            name=data.get("name", "") or data.get("resource_id", ""),
-            label=data.get("label", "") or data.get("accessibility_label", ""),
-            element_type=data.get("element_type", "") or data.get("class_name", ""),
-            clickable=data.get("clickable", True),
-            enabled=data.get("enabled", True),
-            bbox=bbox,
-        )
diff --git a/Agent/platforms/__init__.py b/Agent/platforms/__init__.py
index aacd3ef..25a24a8 100644
--- a/Agent/platforms/__init__.py
+++ b/Agent/platforms/__init__.py
@@ -1,17 +1,15 @@
 from Agent.platforms._mobileconnector import DeviceConnector
-from Agent.platforms._webconnector import WebConnectorRF
 from Agent.platforms._platformfactory import create_platform
-from Agent.platforms.locators import WebLocatorBuilder, MobileLocatorBuilder
-from Agent.platforms.collectors import JSQueryCollector, XMLCollector
+from Agent.platforms.locators import MobileLocatorBuilder
+from Agent.platforms.collectors import XMLCollector
+
+# Placeholder for future web support
+WebConnectorRF = None
 
 __all__ = [
     "DeviceConnector",
     "WebConnectorRF",
     "create_platform",
-    "WebLocatorBuilder",
     "MobileLocatorBuilder",
-    "JSQueryCollector",
     "XMLCollector",
 ]
-
-
diff --git a/Agent/platforms/_mobileconnector.py b/Agent/platforms/_mobileconnector.py
index ca4f700..bc541a7 100644
--- a/Agent/platforms/_mobileconnector.py
+++ b/Agent/platforms/_mobileconnector.py
@@ -86,3 +86,7 @@ def get_screenshot_base64(self) -> str:
     def embed_image_to_log(self, base64_screenshot: str, width: int = 400) -> None:
         msg = f"</td></tr><tr><td colspan=\"3\"><img src=\"data:image/png;base64, {base64_screenshot}\" width=\"{width}\"></td></tr>"
         logger.info(msg, html=True, also_console=False)
+
+    def wait_for_page_stable(self, delay: float = 1.0) -> None:
+        import time
+        time.sleep(delay)
diff --git a/Agent/platforms/_platformfactory.py b/Agent/platforms/_platformfactory.py
index 51bf8f2..2cdbc57 100644
--- a/Agent/platforms/_platformfactory.py
+++ b/Agent/platforms/_platformfactory.py
@@ -1,50 +1,14 @@
-from typing import Union
 from Agent.platforms._mobileconnector import DeviceConnector
-from Agent.platforms._webconnector import WebConnectorRF
 from robot.api import logger
 
-try:
-    from Browser import Browser
-    _has_browser = True
-except ImportError:
-    _has_browser = False
 
-try:
-    from AppiumLibrary import AppiumLibrary
-    _has_appium = True
-except ImportError:
-    _has_appium = False
-
-
-def create_platform(platform_type: str = "auto") -> Union[DeviceConnector, WebConnectorRF]:
+def create_platform(platform_type: str = "auto") -> DeviceConnector:
     """
-    Factory function to create the appropriate platform connector.
-    
-    Args:
-        platform_type: "auto", "web", or "mobile"
-        
-    Returns:
-        Platform connector instance (DeviceConnector or WebConnectorRF)
+    Factory function to create the platform connector.
+    Currently Android-only, web support coming later.
     """
-    if platform_type == "web":
-        logger.info("Creating WebConnectorRF for web automation")
-        return WebConnectorRF()
-    
-    elif platform_type == "mobile":
+    if platform_type == "mobile" or platform_type == "auto":
         logger.info("Creating DeviceConnector for mobile automation")
         return DeviceConnector()
     
-    else:  # auto
-        logger.info("Auto-detecting platform...")
-        
-        if _has_browser:
-            logger.info("Detected Browser library - using WebConnectorRF")
-            return WebConnectorRF()
-        
-        if _has_appium:
-            logger.info("Detected AppiumLibrary - using DeviceConnector")
-            return DeviceConnector()
-        
-        logger.warn("No platform library detected, defaulting to DeviceConnector")
-        return DeviceConnector()
-
+    raise ValueError(f"Platform '{platform_type}' not supported. Use 'mobile' or 'auto'.")
diff --git a/Agent/platforms/_webconnector.py b/Agent/platforms/_webconnector.py
deleted file mode 100644
index 0ada491..0000000
--- a/Agent/platforms/_webconnector.py
+++ /dev/null
@@ -1,59 +0,0 @@
-from typing import Any, Dict, List
-import base64
-from robot.api import logger
-from robot.libraries.BuiltIn import BuiltIn
-from Agent.platforms.collectors.collector_factory import CollectorRegistry
-from Agent.platforms.locators.web import WebLocatorBuilder
-from Agent.ai.prompts.renderer import UIRenderer
-
-
-class WebConnectorRF:
-    """Playwright connector using Robot Framework Browser library keywords."""
-
-    def __init__(self, collector_strategy: str = "js_query") -> None:
-        """
-        Args:
-            collector_strategy: Name of the collector ("js_query")
-        """
-        self.collector_strategy = collector_strategy
-        self.collector = CollectorRegistry.create(collector_strategy)
-        self.locator_builder = WebLocatorBuilder()
-        self._renderer = UIRenderer()
-        logger.debug(f"WebConnectorRF initialized with collector: {collector_strategy}")
-
-    def get_platform(self) -> str:
-        return "web"
-
-    def collect_ui_candidates(self, max_items: int = 500) -> List[Dict[str, Any]]:
-        """Collect interactive web elements."""
-        logger.debug(f"Collecting UI candidates using strategy: {self.collector_strategy}")
-        return self.collector.collect_elements(max_items)
-
-    def build_locator_from_element(self, element: Dict[str, Any]) -> str:
-        """Build Playwright locator from element attributes."""
-        return self.locator_builder.build(element)
-
-    def get_screenshot_base64(self) -> str:
-        """Capture screenshot as base64."""
-        try:
-            screenshot_bytes = BuiltIn().run_keyword('Take Screenshot', 'fullPage')
-            if isinstance(screenshot_bytes, bytes):
-                return base64.b64encode(screenshot_bytes).decode('utf-8')
-            elif isinstance(screenshot_bytes, str):
-                with open(screenshot_bytes, 'rb') as f:
-                    return base64.b64encode(f.read()).decode('utf-8')
-            else:
-                logger.error(f"Unexpected screenshot type: {type(screenshot_bytes)}")
-                return ""
-        except Exception as e:
-            logger.error(f"Failed to capture screenshot: {e}")
-            return ""
-
-    def embed_image_to_log(self, base64_screenshot: str, width: int = 400) -> None:
-        """Embed screenshot into Robot Framework log."""
-        msg = f"</td></tr><tr><td colspan=\"3\"><img src=\"data:image/png;base64, {base64_screenshot}\" width=\"{width}\"></td></tr>"
-        logger.info(msg, html=True, also_console=False)
-
-    def render_ui_for_prompt(self, ui_elements: List[Dict[str, Any]]) -> str:
-        """Render UI elements as text for AI prompt."""
-        return self._renderer.render(ui_elements, platform="web")
diff --git a/Agent/platforms/collectors/__init__.py b/Agent/platforms/collectors/__init__.py
index 1f8b4bf..9d0a93e 100644
--- a/Agent/platforms/collectors/__init__.py
+++ b/Agent/platforms/collectors/__init__.py
@@ -1,25 +1,19 @@
 """
-UI Collectors for web and mobile automation.
+UI Collectors for mobile automation.
 
-This module provides different strategies for collecting UI elements:
-- JSQueryCollector: JavaScript-based querySelector approach (web)
-- XMLCollector: XML page source parsing (mobile)
-
-Use CollectorRegistry to switch between strategies.
+This module provides strategies for collecting UI elements:
+- XMLCollector: XML page source parsing (Android/iOS)
 """
 
 from Agent.platforms.collectors.base_collector import BaseUICollector
 from Agent.platforms.collectors.collector_factory import CollectorRegistry
-from Agent.platforms.collectors.js_query_collector import JSQueryCollector
 from Agent.platforms.collectors.xml_collector import XMLCollector
 from Agent.platforms.collectors.som_renderer import render_som, bbox_center
 
 __all__ = [
     'BaseUICollector',
     'CollectorRegistry',
-    'JSQueryCollector',
     'XMLCollector',
     'render_som',
     'bbox_center',
 ]
-
diff --git a/Agent/platforms/collectors/collector_factory.py b/Agent/platforms/collectors/collector_factory.py
index 0495113..1259391 100644
--- a/Agent/platforms/collectors/collector_factory.py
+++ b/Agent/platforms/collectors/collector_factory.py
@@ -1,8 +1,5 @@
 """
 Factory and Registry for UI Collectors.
-
-This module provides a registry pattern to manage different UI collection
-strategies and switch between them dynamically.
 """
 
 from typing import Dict, List, Type
@@ -11,91 +8,36 @@
 
 
 class CollectorRegistry:
-    """
-    Registry for UI collector strategies.
-    
-    Provides a centralized way to:
-    - Register new collector strategies
-    - Create collector instances by name
-    - List available strategies
-    """
+    """Registry for UI collector strategies."""
     
     _collectors: Dict[str, Type[BaseUICollector]] = {}
     
     @classmethod
     def register(cls, name: str, collector_class: Type[BaseUICollector]) -> None:
-        """
-        Register a new collector strategy.
-        
-        Args:
-            name: Unique identifier for the strategy (e.g., "js_query")
-            collector_class: Class implementing BaseUICollector
-        """
         if not issubclass(collector_class, BaseUICollector):
             raise TypeError(f"{collector_class} must inherit from BaseUICollector")
-        
         cls._collectors[name] = collector_class
         logger.debug(f"Registered UI collector: '{name}' -> {collector_class.__name__}")
     
     @classmethod
     def create(cls, strategy: str) -> BaseUICollector:
-        """
-        Create a collector instance by strategy name.
-        
-        Args:
-            strategy: Name of the registered strategy
-            
-        Returns:
-            Instance of the requested collector
-            
-        Raises:
-            ValueError: If strategy is not registered
-        """
         if strategy not in cls._collectors:
             available = cls.list_available()
-            raise ValueError(
-                f"Unknown collector strategy: '{strategy}'. "
-                f"Available strategies: {available}"
-            )
+            raise ValueError(f"Unknown strategy: '{strategy}'. Available: {available}")
         
         collector_class = cls._collectors[strategy]
-        instance = collector_class()
-        logger.debug(f"Created collector instance: {strategy} ({collector_class.__name__})")
-        return instance
+        return collector_class()
     
     @classmethod
     def list_available(cls) -> List[str]:
-        """
-        List all registered collector strategies.
-        
-        Returns:
-            List of strategy names
-        """
         return list(cls._collectors.keys())
     
     @classmethod
     def is_registered(cls, strategy: str) -> bool:
-        """
-        Check if a strategy is registered.
-        
-        Args:
-            strategy: Name of the strategy
-            
-        Returns:
-            True if registered, False otherwise
-        """
         return strategy in cls._collectors
 
 
-# Auto-register built-in collectors
 def _register_builtin_collectors():
-    """Register all built-in collector strategies."""
-    try:
-        from Agent.platforms.collectors.js_query_collector import JSQueryCollector
-        CollectorRegistry.register("js_query", JSQueryCollector)
-    except ImportError as e:
-        logger.warn(f"Could not register JSQueryCollector: {e}")
-    
     try:
         from Agent.platforms.collectors.xml_collector import XMLCollector
         CollectorRegistry.register("xml", XMLCollector)
@@ -103,6 +45,4 @@ def _register_builtin_collectors():
         logger.warn(f"Could not register XMLCollector: {e}")
 
 
-# Auto-register on module import
 _register_builtin_collectors()
-
diff --git a/Agent/platforms/collectors/js_query_collector.py b/Agent/platforms/collectors/js_query_collector.py
deleted file mode 100644
index 77db9a0..0000000
--- a/Agent/platforms/collectors/js_query_collector.py
+++ /dev/null
@@ -1,142 +0,0 @@
-from typing import Any, Dict, List
-from robot.api import logger
-from robot.libraries.BuiltIn import BuiltIn
-from Agent.platforms.collectors.base_collector import BaseUICollector
-
-
-JS_COLLECT_ELEMENTS = """() => {
-    const selectors = 'button, a, input, select, textarea, label, [type="submit"], [type="button"], [type="reset"], [type="image"], [type="checkbox"], [type="radio"], [type="file"], [type="color"], [type="date"], [type="datetime-local"], [type="email"], [type="month"], [type="number"], [type="range"], [type="search"], [type="tel"], [type="time"], [type="url"], [type="week"], [role="button"], [role="link"], [role="textbox"], [role="checkbox"], [role="radio"], [role="menuitem"], [role="tab"], [role="switch"], [role="slider"], [role="combobox"], [role="listbox"], [role="option"], [role="searchbox"], [role="spinbutton"], [role="menuitemcheckbox"], [role="menuitemradio"], [role="treeitem"], [role="gridcell"], [role="row"], [onclick], [tabindex]:not([tabindex="-1"]), [contenteditable="true"], [contenteditable=""], svg[onclick]';
-    const elements = document.querySelectorAll(selectors);
-    const results = [];
-    elements.forEach(el => {
-        try {
-            const rect = el.getBoundingClientRect();
-            const style = window.getComputedStyle(el);
-            const isVisible = rect.width > 0 && rect.height > 0 && style.visibility !== 'hidden' && style.display !== 'none' && style.opacity !== '0';
-            if (isVisible && !el.disabled && !el.hasAttribute('disabled')) {
-                let labelText = '';
-                const label = el.closest('label') || document.querySelector('label[for="' + el.id + '"]');
-                if (label) {
-                    labelText = label.textContent.replace(el.textContent || '', '').trim().substring(0, 50);
-                }
-                results.push({
-                    text: (el.textContent || el.value || '').trim().substring(0, 100),
-                    id: el.id || '',
-                    className: el.className || '',
-                    name: el.name || el.getAttribute('name') || '',
-                    role: el.role || el.getAttribute('role') || '',
-                    ariaLabel: el.ariaLabel || el.getAttribute('aria-label') || '',
-                    placeholder: el.placeholder || el.getAttribute('placeholder') || '',
-                    label: labelText,
-                    tag: el.tagName.toLowerCase(),
-                    testId: el.getAttribute('data-testid') || '',
-                    type: el.getAttribute('type') || el.type || '',
-                    href: el.href || el.getAttribute('href') || '',
-                    bbox: {
-                        x: Math.round(rect.left),
-                        y: Math.round(rect.top),
-                        width: Math.round(rect.width),
-                        height: Math.round(rect.height)
-                    }
-                });
-            }
-        } catch (e) {}
-    });
-    return results;
-}"""
-
-
-class JSQueryCollector(BaseUICollector):
-    """Collects UI elements using JavaScript querySelector via Playwright."""
-    
-    def __init__(self):
-        self._browser = None
-    
-    def _get_page(self):
-        if self._browser is None:
-            self._browser = BuiltIn().get_library_instance('Browser')
-        return self._browser.get_page()
-    
-    def get_name(self) -> str:
-        return "js_query"
-    
-    def collect_elements(self, max_items: int = 300) -> List[Dict[str, Any]]:
-        """Collect interactive web elements using Playwright evaluate."""
-        logger.debug(f"[{self.get_name()}] Collecting elements via Playwright...")
-        candidates = []
-        
-        try:
-            page = self._get_page()
-            result = page.evaluate(JS_COLLECT_ELEMENTS)
-            logger.debug(f"Playwright returned {len(result) if result else 0} elements")
-            
-            if result and isinstance(result, list):
-                logger.debug(f"[{self.get_name()}] Found {len(result)} raw elements from page")
-                for elem_data in result:
-                    attrs = {
-                        'text': elem_data.get('text', ''),
-                        'resource_id': elem_data.get('testId') or elem_data.get('id', ''),
-                        'aria_label': elem_data.get('ariaLabel', ''),
-                        'placeholder': elem_data.get('placeholder', ''),
-                        'css_class': elem_data.get('className', ''),
-                        'label': elem_data.get('label', ''),
-                        'class_name': elem_data.get('tag', ''),
-                        'role': elem_data.get('role', ''),
-                        'name': elem_data.get('name', ''),
-                        'type': elem_data.get('type', ''),
-                        'href': elem_data.get('href', ''),
-                        'clickable': True,
-                        'enabled': True,
-                        'bbox': elem_data.get('bbox', {}),
-                    }
-                    candidates.append(attrs)
-            else:
-                logger.warn(f"JavaScript did not return expected list: {type(result)}")
-                
-        except Exception as e:
-            logger.error(f"Failed to collect elements via JavaScript: {e}")
-            import traceback
-            logger.debug(f"Traceback: {traceback.format_exc()}")
-        
-        candidates = self._deduplicate_candidates(candidates)
-        candidates.sort(
-            key=lambda x: (
-                # PRIORITY 1: Input fields (text, search, email, etc.) - CRITICAL for forms
-                x.get('class_name') in ['input', 'textarea'],
-                # PRIORITY 2: Elements with placeholder/aria-label (empty inputs)
-                bool(x.get('aria_label')) or bool(x.get('placeholder')),
-                # PRIORITY 3: Elements with visible text
-                bool(x.get('text')) and len(x.get('text', '').strip()) > 0,
-                # PRIORITY 4: Elements with ID or name
-                bool(x.get('resource_id')) or bool(x.get('name')),
-                # PRIORITY 5 (last): Special input types (date, file, color, etc.)
-                x.get('type') not in ['date', 'file', 'color', 'range', 'datetime-local', 'month', 'week', 'time']
-            ),
-            reverse=True
-        )
-        
-        logger.debug(f"[{self.get_name()}] Found {len(candidates)} interactive elements after deduplication")
-        return candidates[:max_items]
-
-    def _deduplicate_candidates(self, candidates: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
-        """Remove duplicate elements based on key attributes."""
-        seen = set()
-        unique = []
-        
-        for candidate in candidates:
-            key = (
-                candidate.get('text', ''),
-                candidate.get('resource_id', ''),
-                candidate.get('aria_label', ''),
-                candidate.get('placeholder', ''),
-                candidate.get('name', ''),
-                candidate.get('class_name', ''),
-                candidate.get('css_class', ''),
-                candidate.get('href', ''),
-            )
-            if key not in seen:
-                seen.add(key)
-                unique.append(candidate)
-        
-        return unique
-
diff --git a/Agent/platforms/collectors/som_renderer.py b/Agent/platforms/collectors/som_renderer.py
index d34827d..ac7a33a 100644
--- a/Agent/platforms/collectors/som_renderer.py
+++ b/Agent/platforms/collectors/som_renderer.py
@@ -44,8 +44,9 @@ def render_som(
     draw = ImageDraw.Draw(overlay)
     
     try:
+        #TODO: fix this for windows and linux ( pixelized font on those OS )
         font = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", 14)
-        font_large = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", 18)
+        font_large = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", 24)
     except:
         font = ImageFont.load_default()
         font_large = font
@@ -66,35 +67,42 @@ def render_som(
         source = element.get(source_key, "dom")
         color = COLOR_DOM if source == "dom" else COLOR_OMNIPARSER if source == "omniparser" else COLOR_DEFAULT
         
+        # Apply margin to create visual spacing between boxes
+        margin = 4
+        box_x1 = x + margin
+        box_y1 = y + margin
+        box_x2 = x + w - margin
+        box_y2 = y + h - margin
+        
         # Draw box with transparency
         draw.rectangle(
-            [x, y, x + w, y + h],
+            [box_x1, box_y1, box_x2, box_y2],
             outline=color + (255,),
             width=2
         )
         
-        # Draw label background
+        # Draw label background (top-left inside box)
         label = str(idx)
         label_bbox = draw.textbbox((0, 0), label, font=font_large)
-        label_w = label_bbox[2] - label_bbox[0] + 8
-        label_h = label_bbox[3] - label_bbox[1] + 4
+        label_w = label_bbox[2] - label_bbox[0] + 12
+        label_h = label_bbox[3] - label_bbox[1] + 8
         
-        label_x = x
-        label_y = y - label_h - 2
-        if label_y < 0:
-            label_y = y + 2
+        label_x = box_x1 + 5
+        label_y = box_y1 + 5
         
         draw.rectangle(
             [label_x, label_y, label_x + label_w, label_y + label_h],
             fill=color + (230,)
         )
         
-        # Draw label text
+        # Draw label text with stroke for better contrast
         draw.text(
-            (label_x + 4, label_y + 2),
+            (label_x + 6, label_y + 4),
             label,
             fill=(255, 255, 255, 255),
-            font=font_large
+            font=font_large,
+            stroke_width=2,
+            stroke_fill=(0, 0, 0, 255)
         )
     
     result = Image.alpha_composite(img, overlay).convert("RGB")
diff --git a/Agent/platforms/locators/__init__.py b/Agent/platforms/locators/__init__.py
index a935906..2bf4991 100644
--- a/Agent/platforms/locators/__init__.py
+++ b/Agent/platforms/locators/__init__.py
@@ -1,5 +1,3 @@
-from Agent.platforms.locators.web import WebLocatorBuilder
 from Agent.platforms.locators.mobile import MobileLocatorBuilder
 
-__all__ = ["WebLocatorBuilder", "MobileLocatorBuilder"]
-
+__all__ = ["MobileLocatorBuilder"]
diff --git a/Agent/platforms/locators/mobile.py b/Agent/platforms/locators/mobile.py
index 96c7fe7..a6fc8d3 100644
--- a/Agent/platforms/locators/mobile.py
+++ b/Agent/platforms/locators/mobile.py
@@ -18,9 +18,9 @@ def build(self, element: Dict[str, Any]) -> str:
     
     def build_android(self, element: Dict[str, Any]) -> str:
         """
-        Build Appium locator for Android.
+        Build XPath locator combining all available attributes.
         
-        Priority: id > accessibility_id > xpath[@text] > class
+        Returns: "//*[@resource-id='x' and @content-desc='y' and @text='z']"
         """
         resource_id = element.get('resource_id', '').strip()
         acc_label = element.get('accessibility_label', '') or element.get('content_desc', '')
@@ -28,25 +28,30 @@ def build_android(self, element: Dict[str, Any]) -> str:
         text = element.get('text', '').strip()
         class_name = element.get('class_name', '').strip()
         
+        conditions = []
+        
         if resource_id:
-            return f"id={resource_id}"
+            conditions.append(f"@resource-id='{resource_id}'")
         
         if acc_label:
-            return f"accessibility_id={acc_label}"
+            conditions.append(f"@content-desc='{acc_label}'")
         
         if text:
-            return f"//*[@text='{text}']"
+            conditions.append(f"@text='{text}'")
         
-        if class_name:
-            return f"class={class_name}"
+        if not conditions:
+            if class_name:
+                return f"//{class_name}"
+            raise AssertionError("Cannot build locator: element has no usable attributes")
         
-        raise AssertionError("Cannot build locator: element has no usable attributes")
+        base = f"//{class_name}" if class_name else "//*"
+        return f"{base}[{' and '.join(conditions)}]"
     
     def build_ios(self, element: Dict[str, Any]) -> str:
         """
-        Build Appium locator for iOS.
+        Build iOS predicate string combining all available attributes.
         
-        Priority: id > accessibility_id > predicate string > class
+        Returns: "-ios predicate string:name == 'x' AND label == 'y'"
         """
         resource_id = element.get('resource_id', '').strip()
         acc_label = element.get('accessibility_label', '') or element.get('label', '')
@@ -54,17 +59,23 @@ def build_ios(self, element: Dict[str, Any]) -> str:
         text = element.get('text', '').strip()
         class_name = element.get('class_name', '').strip()
         
+        conditions = []
+        
         if resource_id:
-            return f"id={resource_id}"
+            conditions.append(f"name == '{resource_id}'")
         
         if acc_label:
-            return f"accessibility_id={acc_label}"
+            escaped = acc_label.replace("'", "\\'")
+            conditions.append(f"label == '{escaped}'")
         
         if text:
-            escaped_text = text.replace("'", "\\'")
-            return f"-ios predicate string:label == '{escaped_text}' OR value == '{escaped_text}'"
+            escaped = text.replace("'", "\\'")
+            conditions.append(f"value == '{escaped}'")
         
         if class_name:
-            return f"class={class_name}"
+            conditions.append(f"type == '{class_name}'")
+        
+        if not conditions:
+            raise AssertionError("Cannot build locator: element has no usable attributes")
         
-        raise AssertionError("Cannot build locator: element has no usable attributes")
+        return f"-ios predicate string:{' AND '.join(conditions)}"
diff --git a/Agent/platforms/locators/web.py b/Agent/platforms/locators/web.py
deleted file mode 100644
index 2f59480..0000000
--- a/Agent/platforms/locators/web.py
+++ /dev/null
@@ -1,82 +0,0 @@
-from typing import Any, Dict
-
-
-class WebLocatorBuilder:
-    """Builds Playwright CSS selectors from element attributes."""
-    
-    def build(self, element: Dict[str, Any]) -> str:
-        """
-        Build Playwright locator combining all non-null attributes.
-        
-        Args:
-            element: Dict with text, role, resource_id, name, class_name, type, href, aria_label, placeholder
-        Returns:
-            CSS/Playwright selector like 'a.sidebar-link[href*="messages"]:text-is("Messages")'
-        """
-        role = element.get('role', '').strip()
-        text = element.get('text', '').strip()
-        aria_label = element.get('aria_label', '').strip()
-        placeholder = element.get('placeholder', '').strip()
-        resource_id = element.get('resource_id', '').strip()
-        name = element.get('name', '').strip()
-        class_name = element.get('class_name', '').strip()
-        css_class = element.get('css_class', '').strip()
-        elem_type = element.get('type', '').strip()
-        href = element.get('href', '').strip()
-        
-        text_cleaned = ' '.join(text.split()) if text else ''
-        
-        css_parts = []
-        
-        if class_name:
-            css_parts.append(class_name)
-        
-        if css_class:
-            first_class = css_class.split()[0]
-            if first_class and not first_class.startswith('_'):
-                css_parts.append(f'.{first_class}')
-        
-        if elem_type and elem_type not in ['text', 'submit', 'button', '']:
-            css_parts.append(f'[type="{elem_type}"]')
-        
-        if name:
-            css_parts.append(f'[name="{name}"]')
-        
-        if resource_id:
-            if 'testid' in resource_id.lower():
-                css_parts.append(f'[data-testid="{resource_id}"]')
-            else:
-                css_parts.append(f'[id="{resource_id}"]')
-        
-        if aria_label:
-            css_parts.append(f'[aria-label="{aria_label}"]')
-        
-        if placeholder:
-            css_parts.append(f'[placeholder="{placeholder}"]')
-        
-        if role:
-            css_parts.append(f'[role="{role}"]')
-        
-        if href and class_name == 'a':
-            if '#' in href:
-                fragment = href.split('#')[-1]
-                if fragment:
-                    css_parts.append(f'[href*="#{fragment}"]')
-            else:
-                path = href.split('/')[-1] if '/' in href else ''
-                if path and len(path) < 60 and '?' not in path:
-                    css_parts.append(f'[href*="{path}"]')
-        
-        css_selector = ''.join(css_parts) if css_parts else ''
-        
-        if text_cleaned and len(text_cleaned) < 80:
-            text_escaped = text_cleaned.replace('\\', '\\\\').replace('"', '\\"')
-            if css_selector:
-                return f'{css_selector}:text-is("{text_escaped}")'
-            return f':text-is("{text_escaped}")'
-        
-        if css_selector:
-            return css_selector
-        
-        raise AssertionError(f"Cannot build locator: no usable attributes in {element}")
-
diff --git a/Agent/platforms/readme.me b/Agent/platforms/readme.me
deleted file mode 100644
index 16a55dc..0000000
--- a/Agent/platforms/readme.me
+++ /dev/null
@@ -1,2 +0,0 @@
-every interaction with mobile ( gettign screenshot , gtting xml , every request 
-from the Device under test ) should be here ( must )
\ No newline at end of file
diff --git a/Agent/tools/base.py b/Agent/tools/base.py
index 912eff6..ef6337f 100644
--- a/Agent/tools/base.py
+++ b/Agent/tools/base.py
@@ -51,29 +51,28 @@ def works_on_locator(self) -> bool:
         return False
     
     @property
-    def works_on_visual(self) -> bool:
-        """Can this tool work with visual detection (coordinates)?
+    def works_on_coordinates(self) -> bool:
+        """Can this tool work with coordinates (x, y)?
         
         Returns:
-            True: Tool can accept visual coordinates (e.g., click, tap)
-            False: Tool cannot work with coordinates alone
+            True: Tool can use coordinates from bounding box
+            False: Tool cannot work with coordinates
         
         Default: False (only action tools need to override)
         """
         return False
     
     @property
-    def has_visual_equivalent(self) -> bool:
-        """Does this tool have a visual equivalent?
+    def has_coordinates_alternative(self) -> bool:
+        """Does this tool have a coordinates-based alternative?
         
         Returns:
-            True: This tool has a visual-based alternative (e.g., tap_element → click_visual_element)
-            False: This tool has no visual equivalent
+            True: This tool has a coordinates-based alternative
+            False: This tool has no coordinates alternative
         
-        Default: False (only tools with visual equivalents need to override)
+        Default: False
         
-        Used for filtering in visual mode: exclude tools with visual equivalents,
-        keep tools without equivalents (even if they only work on locators).
+        Used for filtering in coordinates mode.
         """
         return False
     
diff --git a/Agent/tools/mobile/__init__.py b/Agent/tools/mobile/__init__.py
index 04c9d21..668fada 100644
--- a/Agent/tools/mobile/__init__.py
+++ b/Agent/tools/mobile/__init__.py
@@ -7,7 +7,6 @@
 from Agent.tools.mobile.swipe_up import SwipeUpTool
 from Agent.tools.mobile.hide_keyboard import HideKeyboardTool
 from Agent.tools.mobile.go_back import GoBackTool
-from Agent.tools.mobile.click_visual import ClickVisualElementTool
 
 
 MOBILE_TOOLS = [
@@ -20,7 +19,6 @@
     SwipeUpTool,
     HideKeyboardTool,
     GoBackTool,
-    ClickVisualElementTool,
 ]
 
 __all__ = [
@@ -34,5 +32,4 @@
     "SwipeUpTool",
     "HideKeyboardTool",
     "GoBackTool",
-    "ClickVisualElementTool",
 ]
diff --git a/Agent/tools/mobile/click_element.py b/Agent/tools/mobile/click_element.py
index 73ccf09..c5048f3 100644
--- a/Agent/tools/mobile/click_element.py
+++ b/Agent/tools/mobile/click_element.py
@@ -1,10 +1,17 @@
-from typing import Any, Dict
+from typing import Any, Dict, Tuple
 from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
 from robot.api import logger
 
 
+def get_element_center(element: Dict[str, Any]) -> Tuple[int, int]:
+    bbox = element.get("bbox", {})
+    x = bbox.get("x", 0) + bbox.get("width", 0) // 2
+    y = bbox.get("y", 0) + bbox.get("height", 0) // 2
+    return x, y
+
+
 class ClickElementTool(BaseTool):
-    """Click on a mobile UI element."""
+    """Click on a mobile UI element using coordinates."""
     
     @property
     def name(self) -> str:
@@ -12,7 +19,7 @@ def name(self) -> str:
     
     @property
     def description(self) -> str:
-        return "Click element by INDEX from numbered XML list - USE ONLY when element has clear ID, resource-id, or unique text"
+        return "Tap/click element by INDEX. DO NOT use for text input - use input_text instead."
     
     @property
     def category(self) -> ToolCategory:
@@ -20,15 +27,15 @@ def category(self) -> ToolCategory:
     
     @property
     def works_on_locator(self) -> bool:
-        return True  # Works with XML locator
+        return False
     
     @property
-    def works_on_visual(self) -> bool:
-        return False  # Only works with XML index
+    def works_on_coordinates(self) -> bool:
+        return True
     
     @property
-    def has_visual_equivalent(self) -> bool:
-        return True  # Has click_visual_element as visual alternative
+    def has_coordinates_alternative(self) -> bool:
+        return False
     
     def get_parameters_schema(self) -> Dict[str, Any]:
         return {
@@ -58,8 +65,8 @@ def execute(
             )
         
         element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
+        x, y = get_element_center(element)
         
-        logger.debug(f"Built locator: {locator} from element: {element}")
-        executor.run_keyword("Click Element", locator)
+        logger.debug(f"Tapping at ({x}, {y}) for element: {element.get('text', '')}")
+        executor.run_keyword("Tap", [x, y])
 
diff --git a/Agent/tools/mobile/click_visual.py b/Agent/tools/mobile/click_visual.py
deleted file mode 100644
index c035a4a..0000000
--- a/Agent/tools/mobile/click_visual.py
+++ /dev/null
@@ -1,80 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class ClickVisualElementTool(BaseTool):
-    """Click on element using visual detection (OmniParser).
-    
-    Alternative to tap_element when XML locators are unreliable or unavailable.
-    Uses computer vision (OmniParser) to find elements by visual description.
-    
-    Use cases:
-    - Elements without ID, text, or accessibility labels
-    - Custom UI elements (canvas, images, icons)
-    - Elements not appearing in XML hierarchy
-    - Visual-based interactions
-    """
-    
-    @property
-    def name(self) -> str:
-        return "click_visual_element"
-    
-    @property
-    def description(self) -> str:
-        return "Click element by VISUAL description using computer vision - USE ONLY for icons, images, or elements WITHOUT clear XML identifiers (no ID, no text)"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.MOBILE
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False  # Works with coordinates, not locators
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True  # Designed for visual detection
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the element to click (e.g., 'blue login button', 'profile icon top right')"
-                }
-            },
-            "required": ["element_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual click")
-        
-        logger.debug(f"Visual click: '{element_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Tapping at coordinates: ({x_center}, {y_center})")
-        executor.run_keyword("Tap", [x_center, y_center])
diff --git a/Agent/tools/mobile/go_back.py b/Agent/tools/mobile/go_back.py
index f22a9b1..1a81e1e 100644
--- a/Agent/tools/mobile/go_back.py
+++ b/Agent/tools/mobile/go_back.py
@@ -26,7 +26,7 @@ def works_on_locator(self) -> bool:
         return False  # System navigation action
     
     @property
-    def works_on_visual(self) -> bool:
+    def works_on_coordinates(self) -> bool:
         return False  # System-level action
     
     def get_parameters_schema(self) -> Dict[str, Any]:
diff --git a/Agent/tools/mobile/hide_keyboard.py b/Agent/tools/mobile/hide_keyboard.py
index 0d865e1..fffc99c 100644
--- a/Agent/tools/mobile/hide_keyboard.py
+++ b/Agent/tools/mobile/hide_keyboard.py
@@ -27,7 +27,7 @@ def works_on_locator(self) -> bool:
         return False  # Global device action
     
     @property
-    def works_on_visual(self) -> bool:
+    def works_on_coordinates(self) -> bool:
         return False  # System-level action
     
     def get_parameters_schema(self) -> Dict[str, Any]:
diff --git a/Agent/tools/mobile/input_text.py b/Agent/tools/mobile/input_text.py
index da27955..c6d9679 100644
--- a/Agent/tools/mobile/input_text.py
+++ b/Agent/tools/mobile/input_text.py
@@ -1,10 +1,11 @@
 from typing import Any, Dict
 from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
+from Agent.tools.mobile.click_element import get_element_center
 from robot.api import logger
 
 
 class InputTextTool(BaseTool):
-    """Clear and input text into a mobile UI element."""
+    """Input text into a mobile UI element using coordinates."""
     
     @property
     def name(self) -> str:
@@ -12,7 +13,7 @@ def name(self) -> str:
     
     @property
     def description(self) -> str:
-        return "REQUIRED for text input actions. Input text into a text field using XML element index. Use this tool when instruction contains 'input', 'type', 'enter text', or 'fill' - NEVER use click_visual_element for text input"
+        return "USE THIS when instruction contains 'input', 'type', 'enter', 'fill', 'write', 'saisir', 'taper' or mentions entering text. Types text into a text field."
     
     @property
     def category(self) -> ToolCategory:
@@ -20,11 +21,11 @@ def category(self) -> ToolCategory:
     
     @property
     def works_on_locator(self) -> bool:
-        return True  # Works with XML locator
+        return True
     
     @property
-    def works_on_visual(self) -> bool:
-        return False  # Cannot work with coordinates on iOS
+    def works_on_coordinates(self) -> bool:
+        return True
     
     def get_parameters_schema(self) -> Dict[str, Any]:
         return {
@@ -32,7 +33,7 @@ def get_parameters_schema(self) -> Dict[str, Any]:
             "properties": {
                 "element_index": {
                     "type": "integer",
-                    "description": "The index number of the TEXT FIELD element from the UI elements list (1-based). Find the text field in the numbered list below.",
+                    "description": "The index number of the TEXT FIELD element from the UI elements list (1-based)",
                     "minimum": 1
                 },
                 "text": {
@@ -62,9 +63,9 @@ def execute(
             raise AssertionError("'input_text' requires text argument")
         
         element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
+        x, y = get_element_center(element)
         
-        logger.debug(f"Built locator: {locator}, inputting text: '{text}'")
-        executor.run_keyword("Clear Text", locator)
-        executor.run_keyword("Input Text", locator, text)
-
+        logger.debug(f"Tapping at ({x}, {y}) to focus, then input: '{text}'")
+        executor.run_keyword("Tap", [x, y])
+        executor.run_keyword("Sleep", "1s")
+        executor.run_keyword("Input Text Into Current Element", text)
diff --git a/Agent/tools/mobile/long_press.py b/Agent/tools/mobile/long_press.py
index da3ff98..d3bb67c 100644
--- a/Agent/tools/mobile/long_press.py
+++ b/Agent/tools/mobile/long_press.py
@@ -1,14 +1,11 @@
 from typing import Any, Dict
 from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
+from Agent.tools.mobile.click_element import get_element_center
 from robot.api import logger
 
 
 class LongPressTool(BaseTool):
-    """Long press on a mobile UI element.
-    
-    Uses Appium's Tap keyword with duration parameter to simulate long press.
-    Works on both Android and iOS.
-    """
+    """Long press on a mobile UI element using coordinates."""
     
     @property
     def name(self) -> str:
@@ -16,7 +13,7 @@ def name(self) -> str:
     
     @property
     def description(self) -> str:
-        return "Long press on a mobile UI element (hold for 2 seconds)"
+        return "Long press on element by INDEX (hold for 2 seconds)"
     
     @property
     def category(self) -> ToolCategory:
@@ -24,11 +21,11 @@ def category(self) -> ToolCategory:
     
     @property
     def works_on_locator(self) -> bool:
-        return True  # Works with XML locator
+        return True
     
     @property
-    def works_on_visual(self) -> bool:
-        return False  # AppiumLibrary keyword work on XML locator only
+    def works_on_coordinates(self) -> bool:
+        return True
     
     def get_parameters_schema(self) -> Dict[str, Any]:
         return {
@@ -58,9 +55,8 @@ def execute(
             )
         
         element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
+        x, y = get_element_center(element)
         
-        logger.debug(f"Built locator: {locator}, long pressing for 2s")
-        # Long press = Tap with duration parameter (count=1, duration=2s)
-        executor.run_keyword("Tap", locator, 1, "2s")
+        logger.debug(f"Long pressing at ({x}, {y}) for 2s")
+        executor.run_keyword("Tap", [x, y], 1, "2s")
 
diff --git a/Agent/tools/mobile/scroll_down.py b/Agent/tools/mobile/scroll_down.py
index ea4e1f0..618b67d 100644
--- a/Agent/tools/mobile/scroll_down.py
+++ b/Agent/tools/mobile/scroll_down.py
@@ -23,7 +23,7 @@ def works_on_locator(self) -> bool:
         return False  # Global screen action
     
     @property
-    def works_on_visual(self) -> bool:
+    def works_on_coordinates(self) -> bool:
         return False  # Works on viewport, not specific element
     
     def get_parameters_schema(self) -> Dict[str, Any]:
diff --git a/Agent/tools/mobile/swipe_left.py b/Agent/tools/mobile/swipe_left.py
index 1fae564..0ecb460 100644
--- a/Agent/tools/mobile/swipe_left.py
+++ b/Agent/tools/mobile/swipe_left.py
@@ -26,7 +26,7 @@ def works_on_locator(self) -> bool:
         return False  # Global screen gesture
     
     @property
-    def works_on_visual(self) -> bool:
+    def works_on_coordinates(self) -> bool:
         return False  # Works on viewport, not specific element
     
     def get_parameters_schema(self) -> Dict[str, Any]:
diff --git a/Agent/tools/mobile/swipe_right.py b/Agent/tools/mobile/swipe_right.py
index f87bd2c..8692619 100644
--- a/Agent/tools/mobile/swipe_right.py
+++ b/Agent/tools/mobile/swipe_right.py
@@ -26,7 +26,7 @@ def works_on_locator(self) -> bool:
         return False  # Global screen gesture
     
     @property
-    def works_on_visual(self) -> bool:
+    def works_on_coordinates(self) -> bool:
         return False  # Works on viewport, not specific element
     
     def get_parameters_schema(self) -> Dict[str, Any]:
diff --git a/Agent/tools/mobile/swipe_up.py b/Agent/tools/mobile/swipe_up.py
index 8c0e687..95df12b 100644
--- a/Agent/tools/mobile/swipe_up.py
+++ b/Agent/tools/mobile/swipe_up.py
@@ -4,10 +4,7 @@
 
 
 class SwipeUpTool(BaseTool):
-    """Swipe up on the mobile screen.
-    
-    Useful for scrolling up in lists, closing bottom sheets, or revealing content below.
-    """
+    """Scroll content up on the mobile screen (reveal content above)."""
     
     @property
     def name(self) -> str:
@@ -15,7 +12,7 @@ def name(self) -> str:
     
     @property
     def description(self) -> str:
-        return "Swipe up on the mobile screen (scroll up, close bottom sheet)"
+        return "Scroll content UP (reveal content above)"
     
     @property
     def category(self) -> ToolCategory:
@@ -26,7 +23,7 @@ def works_on_locator(self) -> bool:
         return False  # Global screen gesture
     
     @property
-    def works_on_visual(self) -> bool:
+    def works_on_coordinates(self) -> bool:
         return False  # Works on viewport, not specific element
     
     def get_parameters_schema(self) -> Dict[str, Any]:
@@ -42,6 +39,6 @@ def execute(
         arguments: Dict[str, Any], 
         context: Dict[str, Any]
     ) -> None:
-        # Swipe from bottom (80%) to top (20%) vertically, middle of screen horizontally
-        executor.run_keyword("Swipe By Percent", 50, 80, 50, 20, "1s")
+        # Swipe from top (20%) to bottom (80%) vertically - scrolls content UP
+        executor.run_keyword("Swipe By Percent", 50, 20, 50, 80, "1s")
 
diff --git a/Agent/tools/registry.py b/Agent/tools/registry.py
index 781bc42..b6230e8 100644
--- a/Agent/tools/registry.py
+++ b/Agent/tools/registry.py
@@ -65,30 +65,25 @@ def get_tool_specs(self, category: Optional[Union[ToolCategory, str]] = None) ->
         tools = self.get_by_category(category) if category else self.get_all()
         return [tool.to_tool_spec() for tool in tools]
     
-    def get_tools_for_mode(self, category: Union[ToolCategory, str], mode: str) -> List[BaseTool]:
-        """Get tools filtered by click_mode strategy.
+    def get_tools_for_source(self, category: Union[ToolCategory, str], element_source: str) -> List[BaseTool]:
+        """Get tools filtered by element source.
         
         Args:
-            category: ToolCategory or string ('mobile', 'web', 'visual')
-            mode: 'xml' or 'visual'
-        
-        Returns:
-            Filtered list of tools based on mode
+            category: ToolCategory or string ('mobile')
+            element_source: 'accessibility' or 'vision'
         """
         all_tools = self.get_by_category(category)
         
-        if mode == "visual":
-            # Visual only: Exclude tools that have a visual equivalent (use visual versions)
+        if element_source == "vision":
             return [
                 tool for tool in all_tools
-                if not tool.has_visual_equivalent
+                if not tool.has_coordinates_alternative
             ]
         
-        else:  # xml (default)
-            # XML only: Exclude tools that ONLY work on visual (not on locator)
+        else:
             return [
                 tool for tool in all_tools 
-                if not (tool.works_on_visual and not tool.works_on_locator)
+                if not (tool.works_on_coordinates and not tool.works_on_locator)
             ]
     
     def clear(self) -> None:
diff --git a/Agent/tools/web/__init__.py b/Agent/tools/web/__init__.py
index ee6d644..a22538f 100644
--- a/Agent/tools/web/__init__.py
+++ b/Agent/tools/web/__init__.py
@@ -1,63 +1,2 @@
-from Agent.tools.web.click_element import ClickElementTool
-from Agent.tools.web.input_text import InputTextTool
-from Agent.tools.web.scroll_down import ScrollDownTool
-from Agent.tools.web.scroll_up import ScrollUpTool
-from Agent.tools.web.scroll_to_element import ScrollToElementTool
-from Agent.tools.web.select_option import SelectOptionTool
-from Agent.tools.web.press_key import PressKeyTool
-from Agent.tools.web.go_back import GoBackTool
-from Agent.tools.web.hover import HoverTool
-from Agent.tools.web.double_click import DoubleClickTool
-from Agent.tools.web.clear_text import ClearTextTool
-from Agent.tools.web.click_visual import ClickVisualElementTool
-from Agent.tools.web.input_text_visual import InputTextVisualTool
-from Agent.tools.web.hover_visual import HoverVisualTool
-from Agent.tools.web.double_click_visual import DoubleClickVisualTool
-from Agent.tools.web.select_option_visual import SelectOptionVisualTool
-from Agent.tools.web.scroll_to_element_visual import ScrollToElementVisualTool
-from Agent.tools.web.clear_text_visual import ClearTextVisualTool
-
-
-WEB_TOOLS = [
-    ClickElementTool,
-    InputTextTool,
-    ScrollDownTool,
-    ScrollUpTool,
-    ScrollToElementTool,
-    SelectOptionTool,
-    PressKeyTool,
-    GoBackTool,
-    HoverTool,
-    DoubleClickTool,
-    ClearTextTool,
-    ClickVisualElementTool,
-    InputTextVisualTool,
-    HoverVisualTool,
-    DoubleClickVisualTool,
-    SelectOptionVisualTool,
-    ScrollToElementVisualTool,
-    ClearTextVisualTool,
-]
-
-__all__ = [
-    "WEB_TOOLS",
-    "ClickElementTool",
-    "InputTextTool",
-    "ScrollDownTool",
-    "ScrollUpTool",
-    "ScrollToElementTool",
-    "SelectOptionTool",
-    "PressKeyTool",
-    "GoBackTool",
-    "HoverTool",
-    "DoubleClickTool",
-    "ClearTextTool",
-    "ClickVisualElementTool",
-    "InputTextVisualTool",
-    "HoverVisualTool",
-    "DoubleClickVisualTool",
-    "SelectOptionVisualTool",
-    "ScrollToElementVisualTool",
-    "ClearTextVisualTool",
-]
-
+# Web tools - placeholder for future release
+WEB_TOOLS = []
diff --git a/Agent/tools/web/clear_text.py b/Agent/tools/web/clear_text.py
deleted file mode 100644
index eb9ecc7..0000000
--- a/Agent/tools/web/clear_text.py
+++ /dev/null
@@ -1,65 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class ClearTextTool(BaseTool):
-    """Clear text from an input field."""
-    
-    @property
-    def name(self) -> str:
-        return "clear_text"
-    
-    @property
-    def description(self) -> str:
-        return "Clear text from an input field by index. Use this to empty a text field before typing new text"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the text field to clear from the UI elements list (1-based)",
-                    "minimum": 1
-                }
-            },
-            "required": ["element_index"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator}")
-        executor.run_keyword("Clear Text", locator)
-
diff --git a/Agent/tools/web/clear_text_visual.py b/Agent/tools/web/clear_text_visual.py
deleted file mode 100644
index 845a026..0000000
--- a/Agent/tools/web/clear_text_visual.py
+++ /dev/null
@@ -1,80 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class ClearTextVisualTool(BaseTool):
-    """Clear text from input field using visual detection (OmniParser) for WEB.
-    
-    Finds input field, triple-clicks to select all, and clears.
-    """
-    
-    @property
-    def name(self) -> str:
-        return "clear_text_visual"
-    
-    @property
-    def description(self) -> str:
-        return "Clear text from input field by VISUAL description - USE ONLY when field is NOT in the list or has no clear locator"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the input field to clear"
-                }
-            },
-            "required": ["element_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual clear")
-        
-        logger.debug(f"Visual clear text: '{element_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Clearing text at coordinates: ({x_center}, {y_center})")
-        
-        # Triple-click to select all text
-        executor.run_keyword("Mouse Button", "click", x_center, y_center)
-        executor.run_keyword("Mouse Button", "click", x_center, y_center)
-        executor.run_keyword("Mouse Button", "click", x_center, y_center)
-        
-        # Clear with Backspace
-        executor.run_keyword("Keyboard Input", "press", "Backspace")
diff --git a/Agent/tools/web/click_element.py b/Agent/tools/web/click_element.py
deleted file mode 100644
index e80aaba..0000000
--- a/Agent/tools/web/click_element.py
+++ /dev/null
@@ -1,65 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class ClickElementTool(BaseTool):
-    """Click on a web UI element."""
-    
-    @property
-    def name(self) -> str:
-        return "click_element"
-    
-    @property
-    def description(self) -> str:
-        return "Click element by INDEX from numbered list - USE ONLY when element has clear ID, resource-id, or unique text"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the element from the UI elements list (1-based)",
-                    "minimum": 1
-                }
-            },
-            "required": ["element_index"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator} from element: {element}")
-        executor.run_keyword("Click", locator)
-
diff --git a/Agent/tools/web/click_visual.py b/Agent/tools/web/click_visual.py
deleted file mode 100644
index ee8208c..0000000
--- a/Agent/tools/web/click_visual.py
+++ /dev/null
@@ -1,86 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class ClickVisualElementTool(BaseTool):
-    """Click on element using visual detection (OmniParser) for WEB.
-    
-    Alternative to click_element when locators are unreliable or unavailable.
-    Uses computer vision (OmniParser) to find elements by visual description.
-    """
-    
-    @property
-    def name(self) -> str:
-        return "click_visual_element"
-    
-    @property
-    def description(self) -> str:
-        return "Click element by VISUAL description using computer vision - USE ONLY for icons, images, or elements WITHOUT clear locators (no ID, no text)"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False  # Works with coordinates
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True  # Designed for visual detection
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the element to click (e.g., 'blue login button', 'profile icon top right')"
-                }
-            },
-            "required": ["element_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual click")
-        
-        logger.debug(f"Visual click: '{element_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Clicking at coordinates: ({x_center}, {y_center})")
-        # Use 'Click With Options' or 'Mouse Button' - Click With Options is more versatile
-        # We need to pass coordinates. Browser library 'Click' takes a selector.
-        # To click at coordinates, we use 'Mouse Button' or 'Click' with position.
-        # Actually, 'Click' keyword in Browser library takes a selector.
-        # To click at coordinates, we can use 'Mouse Button' keyword with x, y arguments if available,
-        # OR we can use 'Click' on 'body' with offset.
-        # Let's check 'Click With Options' documentation or usage in other tools.
-        # Wait, 'Click' in Playwright/Browser usually targets an element.
-        # To click at x,y, we usually do: page.mouse.click(x, y).
-        # In Robot Framework Browser library:
-        # 'Mouse Button' keyword: Mouse Button    action=click    x=100    y=200
-        
-        executor.run_keyword("Mouse Button", "click", x_center, y_center)
diff --git a/Agent/tools/web/double_click.py b/Agent/tools/web/double_click.py
deleted file mode 100644
index c321105..0000000
--- a/Agent/tools/web/double_click.py
+++ /dev/null
@@ -1,65 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class DoubleClickTool(BaseTool):
-    """Double click an element."""
-    
-    @property
-    def name(self) -> str:
-        return "double_click"
-    
-    @property
-    def description(self) -> str:
-        return "Double click an element by index. Use this when a single click is not enough"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the element to double click from the UI elements list (1-based)",
-                    "minimum": 1
-                }
-            },
-            "required": ["element_index"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator}")
-        executor.run_keyword("Click With Options", locator, "left", "clickCount=2")
-
diff --git a/Agent/tools/web/double_click_visual.py b/Agent/tools/web/double_click_visual.py
deleted file mode 100644
index 0bd7e34..0000000
--- a/Agent/tools/web/double_click_visual.py
+++ /dev/null
@@ -1,74 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class DoubleClickVisualTool(BaseTool):
-    """Double click on element using visual detection (OmniParser) for WEB.
-    
-    Alternative to double_click when locators are unreliable.
-    Uses computer vision to find the element and double clicks it.
-    """
-    
-    @property
-    def name(self) -> str:
-        return "double_click_visual"
-    
-    @property
-    def description(self) -> str:
-        return "Double click element by VISUAL description - USE ONLY when standard double_click fails or for elements without clear locators"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the element to double click"
-                }
-            },
-            "required": ["element_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual double click")
-        
-        logger.debug(f"Visual double click: '{element_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Double clicking at coordinates: ({x_center}, {y_center})")
-        
-        executor.run_keyword("Mouse Button", "click", x_center, y_center, "clickCount=2")
diff --git a/Agent/tools/web/go_back.py b/Agent/tools/web/go_back.py
deleted file mode 100644
index 6dec212..0000000
--- a/Agent/tools/web/go_back.py
+++ /dev/null
@@ -1,43 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class GoBackTool(BaseTool):
-    """Navigate back in browser history (like pressing browser back button)."""
-    
-    @property
-    def name(self) -> str:
-        return "go_back"
-    
-    @property
-    def description(self) -> str:
-        return "Go back one step in browser history (back button). Use this to navigate to the previous page"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {},
-            "required": []
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        executor.run_keyword("Go Back")
-
diff --git a/Agent/tools/web/hover.py b/Agent/tools/web/hover.py
deleted file mode 100644
index 43e0e7e..0000000
--- a/Agent/tools/web/hover.py
+++ /dev/null
@@ -1,65 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class HoverTool(BaseTool):
-    """Hover over an element."""
-    
-    @property
-    def name(self) -> str:
-        return "hover"
-    
-    @property
-    def description(self) -> str:
-        return "Hover over an element by index. Use this to reveal hidden menus, tooltips, or trigger hover effects"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the element to hover over from the UI elements list (1-based)",
-                    "minimum": 1
-                }
-            },
-            "required": ["element_index"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator}")
-        executor.run_keyword("Hover", locator)
-
diff --git a/Agent/tools/web/hover_visual.py b/Agent/tools/web/hover_visual.py
deleted file mode 100644
index a923b83..0000000
--- a/Agent/tools/web/hover_visual.py
+++ /dev/null
@@ -1,74 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class HoverVisualTool(BaseTool):
-    """Hover over element using visual detection (OmniParser) for WEB.
-    
-    Alternative to hover when locators are unreliable.
-    Uses computer vision to find the element and moves mouse cursor to it.
-    """
-    
-    @property
-    def name(self) -> str:
-        return "hover_visual"
-    
-    @property
-    def description(self) -> str:
-        return "Hover over element by VISUAL description - USE ONLY when standard hover fails or for elements without clear locators"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the element to hover over"
-                }
-            },
-            "required": ["element_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual hover")
-        
-        logger.debug(f"Visual hover: '{element_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Hovering at coordinates: ({x_center}, {y_center})")
-        executor.run_keyword("Mouse Move", x_center, y_center)
diff --git a/Agent/tools/web/input_text.py b/Agent/tools/web/input_text.py
deleted file mode 100644
index b99f003..0000000
--- a/Agent/tools/web/input_text.py
+++ /dev/null
@@ -1,92 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class InputTextTool(BaseTool):
-    """Clear and input text into a web UI element."""
-    
-    @property
-    def name(self) -> str:
-        return "input_text"
-    
-    @property
-    def description(self) -> str:
-        return "REQUIRED for text input actions. Input text into a text field using element index. Use this tool when instruction contains 'input', 'type', 'enter text', or 'fill'"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the TEXT FIELD element from the UI elements list (1-based). Find the text field in the numbered list below.",
-                    "minimum": 1
-                },
-                "text": {
-                    "type": "string",
-                    "description": "The text to input into the element"
-                }
-            },
-            "required": ["element_index", "text"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        text = arguments["text"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        if not text:
-            raise AssertionError("'input_text' requires text argument")
-        
-        element = ui_candidates[element_index - 1]
-        
-        class_name = element.get('class_name', '')
-        elem_type = element.get('type', '')
-        
-        valid_input_elements = ['input', 'textarea']
-        valid_input_types = ['text', 'search', 'email', 'password', 'tel', 'url', 'number', '']
-        
-        if class_name not in valid_input_elements:
-            raise AssertionError(
-                f"Cannot input text: element at index {element_index} is a '{class_name}', not a text input field. "
-                f"Expected: input or textarea. Element: {element}"
-            )
-        
-        if class_name == 'input' and elem_type and elem_type not in valid_input_types:
-            raise AssertionError(
-                f"Cannot input text: input element at index {element_index} has type '{elem_type}' which doesn't accept text. "
-                f"Expected types: {valid_input_types}. Element: {element}"
-            )
-        
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator}, inputting text: '{text}'")
-        executor.run_keyword("Fill Text", locator, text)
-
diff --git a/Agent/tools/web/input_text_visual.py b/Agent/tools/web/input_text_visual.py
deleted file mode 100644
index f8b1f88..0000000
--- a/Agent/tools/web/input_text_visual.py
+++ /dev/null
@@ -1,84 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class InputTextVisualTool(BaseTool):
-    """Input text into element using visual detection (OmniParser) for WEB.
-    
-    Alternative to input_text when locators are unreliable.
-    Uses computer vision to find the input field and types text.
-    """
-    
-    @property
-    def name(self) -> str:
-        return "input_text_visual"
-    
-    @property
-    def description(self) -> str:
-        return "Input text into element by VISUAL description - USE ONLY when standard input_text fails or for elements without clear locators"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the input field (e.g., 'search bar at top', 'email input field')"
-                },
-                "text": {
-                    "type": "string",
-                    "description": "The text to input"
-                }
-            },
-            "required": ["element_description", "text"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        text = arguments["text"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual input")
-        
-        logger.debug(f"Visual input: '{element_description}' -> '{text}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Clicking to focus at coordinates: ({x_center}, {y_center})")
-        # Click to focus
-        executor.run_keyword("Mouse Button", "click", x_center, y_center)
-        
-        # Type text
-        logger.debug(f"Typing text: '{text}'")
-        executor.run_keyword("Keyboard Input", "type", text)
diff --git a/Agent/tools/web/press_key.py b/Agent/tools/web/press_key.py
deleted file mode 100644
index 4a7aa8c..0000000
--- a/Agent/tools/web/press_key.py
+++ /dev/null
@@ -1,53 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class PressKeyTool(BaseTool):
-    """Press a keyboard key (Enter, Escape, Tab, Arrow keys, etc.)."""
-    
-    @property
-    def name(self) -> str:
-        return "press_key"
-    
-    @property
-    def description(self) -> str:
-        return "Press a keyboard key. Common keys: Enter, Escape, Tab, ArrowUp, ArrowDown, ArrowLeft, ArrowRight, Backspace, Delete. Use this to submit forms (Enter), close modals (Escape), navigate (Arrow keys), or clear input (Backspace)"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "key": {
-                    "type": "string",
-                    "description": "The key to press. Common values: Enter, Escape, Tab, ArrowUp, ArrowDown, ArrowLeft, ArrowRight, Backspace, Delete, Space",
-                    "enum": ["Enter", "Escape", "Tab", "ArrowUp", "ArrowDown", "ArrowLeft", "ArrowRight", "Backspace", "Delete", "Space"]
-                }
-            },
-            "required": ["key"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        key = arguments["key"]
-        
-        logger.debug(f"Pressing key: {key}")
-        # Browser library: Press Keys selector *keys
-        executor.run_keyword("Press Keys", "body", key)
-
diff --git a/Agent/tools/web/scroll_down.py b/Agent/tools/web/scroll_down.py
deleted file mode 100644
index 3275a85..0000000
--- a/Agent/tools/web/scroll_down.py
+++ /dev/null
@@ -1,43 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class ScrollDownTool(BaseTool):
-    """Scroll down the web page."""
-    
-    @property
-    def name(self) -> str:
-        return "scroll_down"
-    
-    @property
-    def description(self) -> str:
-        return "Scroll down the web page by one viewport height. Use this for general scrolling down to see more content. For scrolling to a specific element (like footer, button, etc.), use scroll_to_element instead"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {},
-            "required": []
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        executor.run_keyword("Scroll By", None, "height")
-
diff --git a/Agent/tools/web/scroll_to_element.py b/Agent/tools/web/scroll_to_element.py
deleted file mode 100644
index 1b29a06..0000000
--- a/Agent/tools/web/scroll_to_element.py
+++ /dev/null
@@ -1,65 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class ScrollToElementTool(BaseTool):
-    """Scroll to a specific element on the page."""
-    
-    @property
-    def name(self) -> str:
-        return "scroll_to_element"
-    
-    @property
-    def description(self) -> str:
-        return "Scroll the page to bring a specific element into view by index. Use this when you need to scroll to a specific element like footer, header, button, or any other named element. Examples: 'scroll to footer', 'scroll to submit button', 'scroll to contact section'"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the element to scroll to from the UI elements list (1-based)",
-                    "minimum": 1
-                }
-            },
-            "required": ["element_index"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator}")
-        executor.run_keyword("Scroll To Element", locator)
-
diff --git a/Agent/tools/web/scroll_to_element_visual.py b/Agent/tools/web/scroll_to_element_visual.py
deleted file mode 100644
index a432a58..0000000
--- a/Agent/tools/web/scroll_to_element_visual.py
+++ /dev/null
@@ -1,76 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class ScrollToElementVisualTool(BaseTool):
-    """Scroll to element using visual detection (OmniParser) for WEB.
-    
-    Finds element coordinates and scrolls to it using JavaScript.
-    """
-    
-    @property
-    def name(self) -> str:
-        return "scroll_to_element_visual"
-    
-    @property
-    def description(self) -> str:
-        return "Scroll to element by VISUAL description - USE ONLY when element is NOT in the list or has no clear locator"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_description": {
-                    "type": "string",
-                    "description": "Visual description of the element to scroll to"
-                }
-            },
-            "required": ["element_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_description = arguments["element_description"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual scroll")
-        
-        logger.debug(f"Visual scroll to: '{element_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        result = orchestrator.find_element(
-            element_description=element_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Element '{element_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Scrolling to coordinates: ({x_center}, {y_center})")
-        
-        # Use JavaScript to scroll to coordinates
-        scroll_js = f"window.scrollTo({{top: {y_center} - window.innerHeight / 2, behavior: 'smooth'}});"
-        executor.run_keyword("Evaluate JavaScript", "body >> nth=0", scroll_js)
diff --git a/Agent/tools/web/scroll_up.py b/Agent/tools/web/scroll_up.py
deleted file mode 100644
index 8f42c67..0000000
--- a/Agent/tools/web/scroll_up.py
+++ /dev/null
@@ -1,43 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class ScrollUpTool(BaseTool):
-    """Scroll up the web page."""
-    
-    @property
-    def name(self) -> str:
-        return "scroll_up"
-    
-    @property
-    def description(self) -> str:
-        return "Scroll up the web page by one viewport height. Use this for general scrolling up to see previous content. For scrolling to a specific element at the top, use scroll_to_element instead"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {},
-            "required": []
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        executor.run_keyword("Scroll By", None, "-height")
-
diff --git a/Agent/tools/web/select_option.py b/Agent/tools/web/select_option.py
deleted file mode 100644
index 73d54fd..0000000
--- a/Agent/tools/web/select_option.py
+++ /dev/null
@@ -1,73 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from robot.api import logger
-
-
-class SelectOptionTool(BaseTool):
-    """Select an option from a dropdown/select element."""
-    
-    @property
-    def name(self) -> str:
-        return "select_option"
-    
-    @property
-    def description(self) -> str:
-        return "Select an option from a dropdown/select element by index. Use this for <select> elements, dropdowns, or any element with selectable options"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return True
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return False
-    
-    @property
-    def has_visual_equivalent(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "element_index": {
-                    "type": "integer",
-                    "description": "The index number of the SELECT/DROPDOWN element from the UI elements list (1-based)",
-                    "minimum": 1
-                },
-                "option_value": {
-                    "type": "string",
-                    "description": "The value or text of the option to select. Can be the option text or value attribute"
-                }
-            },
-            "required": ["element_index", "option_value"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        element_index = arguments["element_index"]
-        option_value = arguments["option_value"]
-        ui_candidates = context.get("ui_candidates", [])
-        
-        if element_index < 1 or element_index > len(ui_candidates):
-            raise AssertionError(
-                f"Invalid element_index: {element_index}. Must be 1-{len(ui_candidates)}"
-            )
-        
-        if not option_value:
-            raise AssertionError("'select_option' requires option_value argument")
-        
-        element = ui_candidates[element_index - 1]
-        locator = executor.build_locator(element)
-        
-        logger.debug(f"Built locator: {locator}, selecting option: {option_value}")
-        executor.run_keyword("Select Options By", locator, "text", option_value)
-
diff --git a/Agent/tools/web/select_option_visual.py b/Agent/tools/web/select_option_visual.py
deleted file mode 100644
index dbc7a97..0000000
--- a/Agent/tools/web/select_option_visual.py
+++ /dev/null
@@ -1,101 +0,0 @@
-from typing import Any, Dict
-from Agent.tools.base import BaseTool, ExecutorProtocol, ToolCategory
-from Agent.ai.vlm.interface import OmniParserOrchestrator
-from robot.api import logger
-
-
-class SelectOptionVisualTool(BaseTool):
-    """Select option from dropdown using visual detection (OmniParser) for WEB.
-    
-    Two-step process:
-    1. Click dropdown by visual description
-    2. Click option by visual description
-    """
-    
-    @property
-    def name(self) -> str:
-        return "select_option_visual"
-    
-    @property
-    def description(self) -> str:
-        return "Select option from dropdown by VISUAL description - USE ONLY when dropdown is NOT in the list or has no clear locator"
-    
-    @property
-    def category(self) -> ToolCategory:
-        return ToolCategory.WEB
-    
-    @property
-    def works_on_locator(self) -> bool:
-        return False
-    
-    @property
-    def works_on_visual(self) -> bool:
-        return True
-    
-    def get_parameters_schema(self) -> Dict[str, Any]:
-        return {
-            "type": "object",
-            "properties": {
-                "dropdown_description": {
-                    "type": "string",
-                    "description": "Visual description of the dropdown/select element"
-                },
-                "option_description": {
-                    "type": "string",
-                    "description": "Visual description of the option to select"
-                }
-            },
-            "required": ["dropdown_description", "option_description"]
-        }
-    
-    def execute(
-        self, 
-        executor: ExecutorProtocol, 
-        arguments: Dict[str, Any], 
-        context: Dict[str, Any]
-    ) -> None:
-        dropdown_description = arguments["dropdown_description"]
-        option_description = arguments["option_description"]
-        
-        # Get screenshot from context
-        screenshot_base64 = context.get("screenshot_base64")
-        
-        if not screenshot_base64:
-            raise AssertionError("screenshot_base64 not provided in context for visual select")
-        
-        logger.debug(f"Visual select: dropdown='{dropdown_description}', option='{option_description}'")
-        
-        orchestrator = OmniParserOrchestrator()
-        
-        # Step 1: Find and click dropdown
-        result = orchestrator.find_element(
-            element_description=dropdown_description,
-            image_base64=screenshot_base64
-        )
-        
-        if not result:
-            raise AssertionError(f"Dropdown '{dropdown_description}' not found")
-        
-        x_center, y_center = OmniParserOrchestrator.get_element_center_coordinates(result)
-        
-        logger.debug(f"Clicking dropdown at coordinates: ({x_center}, {y_center})")
-        executor.run_keyword("Mouse Button", "click", x_center, y_center)
-        
-        # Wait for dropdown to open
-        executor.run_keyword("Sleep", "0.5s")
-        
-        # Step 2: Capture new screenshot and find option
-        screenshot_base64_2 = executor.platform.get_screenshot_base64()
-        
-        result_option = orchestrator.find_element(
-            element_description=option_description,
-            image_base64=screenshot_base64_2
-        )
-        
-        if not result_option:
-            raise AssertionError(f"Option '{option_description}' not found")
-        
-        x_opt, y_opt = OmniParserOrchestrator.get_element_center_coordinates(result_option)
-        
-        logger.debug(f"Clicking option at coordinates: ({x_opt}, {y_opt})")
-        executor.run_keyword("Mouse Button", "click", x_opt, y_opt)
diff --git a/Agent/utilities/_logdir.py b/Agent/utilities/_logdir.py
new file mode 100644
index 0000000..aeb01ab
--- /dev/null
+++ b/Agent/utilities/_logdir.py
@@ -0,0 +1,31 @@
+
+import os
+from pathlib import Path
+from robot.libraries.BuiltIn import BuiltIn
+
+def get_artifacts_folder_name() -> str:
+    """Returns name of the folder where Robot logs/reports are generated."""
+    variables = BuiltIn().get_variables()
+    robot_output_dir = str(variables['${OUTPUTDIR}'])
+    exec_dir = str(variables['${EXECDIR}'])
+
+    if 'pabot_results' in robot_output_dir:
+        robot_output_dir = robot_output_dir.split('pabot_results')[0]
+
+    relative_path = str(robot_output_dir).replace(str(exec_dir), '').strip('/')
+    output_folder = relative_path.split('/')[0]
+    return output_folder
+
+def get_artifacts_root_path() -> Path:
+    """Full path to artifacts root dir. Example: /path/to/results"""
+    variables = BuiltIn().get_variables()
+    exec_dir = str(variables['${EXECDIR}'])
+    artifacts_folder = get_artifacts_folder_name()
+    return Path(exec_dir) / artifacts_folder
+
+def set_artifacts_subdir(ArtifactsName) -> Path:
+    """Returns full path to a subdir in artifacts root (creates if missing)."""
+    artifacts_dir = get_artifacts_root_path() / ArtifactsName
+    artifacts_dir.mkdir(parents=True, exist_ok=True)
+    return artifacts_dir
+
diff --git a/Agent/utilities/imguploader/_imgbb.py b/Agent/utilities/imguploader/_imgbb.py
index 63bde4e..32eb07a 100644
--- a/Agent/utilities/imguploader/_imgbb.py
+++ b/Agent/utilities/imguploader/_imgbb.py
@@ -1,6 +1,5 @@
 import requests
 from typing import Optional
-from robot.api import logger
 from Agent.config.config import Config
 from Agent.utilities.imguploader._imgbase import BaseImageUploader
 
@@ -13,32 +12,18 @@ def __init__(self):
 
     @property
     def api_key(self):
-        api_key = self.config.IMGBB_API_KEY
-        if not api_key:
-            logger.error("IMGBB_API_KEY not found in configuration")
-        return api_key
+        return self.config.IMGBB_API_KEY
 
     def _make_request(self, payload: dict, files: bool = False) -> Optional[str]:
         try:
             if files:
-                response = requests.post(self.base_url, files=payload)
+                response = requests.post(self.base_url, files=payload, timeout=10)
             else:
-                response = requests.post(self.base_url, data=payload, headers=self.headers)
-            
-            if not response.ok:
-                logger.error(f"ImgBB API Error: {response.status_code} - {response.text}")
-                
+                response = requests.post(self.base_url, data=payload, headers=self.headers, timeout=10)
             response.raise_for_status()
             json_data = response.json()
             return self._extract_url(json_data)
-        except requests.exceptions.RequestException as e:
-            logger.error(f"API Request Failed: {e}")
-            return None
-        except ValueError:
-            logger.error("Invalid JSON response from API")
-            return None
-        except Exception as e:
-            logger.error(f"Unexpected error during image upload: {str(e)}")
+        except Exception:
             return None
 
     def _extract_url(self, json_data: dict) -> Optional[str]:
diff --git a/Agent/utilities/imguploader/_imghost.py b/Agent/utilities/imguploader/_imghost.py
index 98e6a17..f38291b 100644
--- a/Agent/utilities/imguploader/_imghost.py
+++ b/Agent/utilities/imguploader/_imghost.py
@@ -2,7 +2,6 @@
 import requests
 from Agent.config.config import Config
 from Agent.utilities.imguploader._imgbase import BaseImageUploader
-from robot.api import logger
 
 
 class FreeImageHostUploader(BaseImageUploader):
@@ -13,28 +12,18 @@ def __init__(self):
 
     @property
     def api_key(self):
-        api_key = self.config.FREEIMAGEHOST_API_KEY
-        if not api_key:
-            logger.error("FREEIMAGEHOST_API_KEY not found in configuration")
-        return api_key
+        return self.config.FREEIMAGEHOST_API_KEY
 
     def _make_request(self, payload: dict, files: bool = False) -> Optional[str]:
         try:
             if files:
-                response = requests.post(self.base_url, files=payload)
+                response = requests.post(self.base_url, files=payload, timeout=10)
             else:
-                response = requests.post(self.base_url, data=payload, headers=self.headers)
+                response = requests.post(self.base_url, data=payload, headers=self.headers, timeout=10)
             response.raise_for_status()
             json_data = response.json()
             return self._extract_url(json_data)
-        except requests.exceptions.RequestException as e:
-            logger.error(f"From image host uploader: API Request Failed: {e}")
-            return None
-        except ValueError:
-            logger.error("From image host uploader: Invalid JSON response from API")
-            return None
-        except Exception as e:
-            logger.error(f"From image host uploader: Other unexpected error: {e}")
+        except Exception:
             return None
 
     def _extract_url(self, json_data: dict) -> Optional[str]:
diff --git a/Agent/utilities/imguploader/imghandler.py b/Agent/utilities/imguploader/imghandler.py
index 693d23f..92f6e05 100644
--- a/Agent/utilities/imguploader/imghandler.py
+++ b/Agent/utilities/imguploader/imghandler.py
@@ -24,55 +24,34 @@ def upload_from_base64(self, base64_data: str) -> Optional[str]:
         Returns base64 if all providers fail or none configured.
         """
         if not self.uploaders:
-            logger.warn("Fallback: returning the image in base64 (no provider configured)")
+            logger.debug("No upload provider configured, using base64")
             return f"data:image/png;base64,{base64_data}"
         
-        # Try each provider in order
-        for idx, uploader in enumerate(self.uploaders):
+        for uploader in self.uploaders:
             provider_name = type(uploader).__name__
-            logger.debug(f"Attempting upload with {provider_name} ({idx+1}/{len(self.uploaders)})")
             try:
                 result = uploader.upload_from_base64(base64_data)
-                
                 if result:
-                    logger.info(f"Image uploaded successfully with {provider_name}")
+                    logger.debug(f"Image uploaded via {provider_name}")
                     return result
-                else:
-                    logger.warn(f"{provider_name} failed to upload (returned None)")
-                    
-            except Exception as e:
-                logger.warn(f"{provider_name} raised an error: {str(e)}")
+            except Exception:
+                pass
         
-        logger.warn("Fallback: returning the image in base64 (all providers failed)")
+        logger.debug("Upload failed, using base64 fallback")
         return f"data:image/png;base64,{base64_data}"
 
     def _build_uploaders_list(self, service: str) -> list[BaseImageUploader]:
-        """
-        Builds list of uploaders based on service selection.
-        In auto mode, adds all available providers (imgbb has priority).
-        """
         uploaders = []
         
-        # Debug config loading
-        has_imgbb = bool(self.config.IMGBB_API_KEY)
-        has_freeimage = bool(self.config.FREEIMAGEHOST_API_KEY)
-        logger.debug(f"ImageUploader init: service={service}, has_imgbb={has_imgbb}, has_freeimage={has_freeimage}")
-        
         if service == "imgbb" and self.config.IMGBB_API_KEY:
             uploaders.append(ImgBBUploader())
         elif service == "freeimagehost" and self.config.FREEIMAGEHOST_API_KEY:
             uploaders.append(FreeImageHostUploader())
         elif service == "auto":
-            # Priority order: imgbb first, then freeimagehost as fallback
             if self.config.IMGBB_API_KEY:
                 uploaders.append(ImgBBUploader())
             if self.config.FREEIMAGEHOST_API_KEY:
                 uploaders.append(FreeImageHostUploader())
         
-        if not uploaders:
-            logger.warn("No upload service configured. Images will be returned in base64.")
-        else:
-            logger.debug(f"Configured uploaders: {[type(u).__name__ for u in uploaders]}")
-        
         return uploaders
 
diff --git a/Agent/utilities/imguploader/readme.me b/Agent/utilities/imguploader/readme.me
deleted file mode 100644
index 64369b1..0000000
--- a/Agent/utilities/imguploader/readme.me
+++ /dev/null
@@ -1,14 +0,0 @@
-
-
-imghandler is the facade and have one responsability : 
-recieve image base64 return link or base64 if failed to upload 
-
-whatever the api will be used ( there is a default one , but another ones for backup )
-
-
-current working which is enough for mvp :
-
-Handles 3 fallback cases with warnings:
-✅ No provider configured → returns base64 + warning
-✅ Upload fails (returns None) → returns base64 + warning
-✅ Exception raised → returns base64 + warning
diff --git a/MANIFEST.in b/MANIFEST.in
index 1da6c93..98ec05b 100644
--- a/MANIFEST.in
+++ b/MANIFEST.in
@@ -1,21 +1,17 @@
-# Include the license file
 include LICENSE
 
-# Include the readme
 include readme.md
+include CONTRIBUTING.md
+
+include pyproject.toml
+include env.example
 
-# Include configuration files
 include Agent/config/*.json
 
-# Exclude test data and other non-essential files
-exclude tests/*
-exclude atest_agent.robot
-exclude browserstack.yml
-exclude *.log
-exclude *.html
-exclude *.xml
+prune Reports
+prune browser
+prune tests
 
-# Exclude virtual environment and cache
 global-exclude __pycache__
 global-exclude *.py[co]
 global-exclude .DS_Store
diff --git a/Makefile b/Makefile
index 909b5f3..817443f 100644
--- a/Makefile
+++ b/Makefile
@@ -1,51 +1,50 @@
-# Makefile pour le projet Robot Framework Agent
 
 .PHONY: help install install-dev test lint format clean
 
-help:  ## Afficher cette aide
+help:
 	@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-20s\033[0m %s\n", $$1, $$2}'
 
-install:  ## Installer en mode production
+install:
 	pip install .
 
-install-dev:  ## Installer en mode développement
+install-dev:
 	pip install -e ".[dev]"
 	pre-commit install
 
-test:  ## Lancer les tests
+test:
 	pytest tests/utest/ -v
 
-test-cov:  ## Lancer les tests avec coverage
+test-cov:
 	pytest tests/utest/ --cov=Agent --cov-report=html --cov-report=term
 
-lint:  ## Vérifier le code (flake8 + mypy)
+lint:
 	flake8 Agent/ tests/
 	mypy Agent/ --ignore-missing-imports
 
-format:  ## Formater le code (black + isort)
+format:
 	black Agent/ tests/
 	isort Agent/ tests/
 
-format-check:  ## Vérifier le formatage sans modifier
+format-check:
 	black --check Agent/ tests/
 	isort --check Agent/ tests/
 
-clean:  ## Nettoyer les fichiers générés
+clean:
 	rm -rf build/ dist/ *.egg-info
 	rm -rf .pytest_cache/ .coverage htmlcov/
 	rm -rf Agent/__pycache__/ Agent/**/__pycache__/
 	find . -type d -name __pycache__ -exec rm -rf {} +
 	find . -type f -name "*.pyc" -delete
 
-build:  ## Construire le package
+build:
 	python -m build
 
-publish-test:  ## Publier sur TestPyPI
+publish-test:
 	twine upload --repository testpypi dist/*
 
-publish:  ## Publier sur PyPI
+publish:
 	twine upload dist/*
 
-pre-commit:  ## Lancer pre-commit sur tous les fichiers
+pre-commit:
 	pre-commit run --all-files
 
diff --git a/demo_android.robot b/demo_android.robot
new file mode 100644
index 0000000..205b59d
--- /dev/null
+++ b/demo_android.robot
@@ -0,0 +1,53 @@
+*** Settings ***
+Documentation    Demo: AI-driven Android test with Robot Framework Agent
+...              Standalone file - just update the variables below and run
+Library    AppiumLibrary
+Library    Agent    llm_client=openai    llm_model=gpt-4.1    platform_type=mobile    element_source=accessibility
+
+
+*** Variables ***
+# Appium server
+${APPIUM_URL}              http://127.0.0.1:4723
+
+# Device config
+${DEVICE_NAME}             YOUR_DEVICE_NAME
+${PLATFORM_NAME}           Android
+${AUTOMATION_NAME}         UiAutomator2
+
+# App config (choose one)
+${APP_PACKAGE}             com.google.android.youtube
+${APP_ACTIVITY}            .app.honeycomb.Shell$HomeActivity
+
+
+*** Test Cases ***
+Demo Youtube
+    [Documentation]    Demo Youtube
+    Open Demo App
+
+    Agent.check    instruction=Verify that YouTube watch history is turned off
+
+    Agent.Do       instruction=Tap the search icon at the top
+
+    Agent.Do       instruction=Enter "Robot Framework" into the search input box
+
+    Agent.check    instruction=Confirm every result displayed includes "Robot Framework" in its title or description
+
+    Agent.Do       instruction=Click on the first search result appearing in the search results list
+
+    Agent.Ask      question=How many subscribers does the Robot Framework channel have?
+
+
+
+
+*** Keywords ***
+Open Demo App
+    [Documentation]    Opens the configured Android app
+    Open Application    
+    ...    remote_url=${APPIUM_URL}
+    ...    platformName=${PLATFORM_NAME}
+    ...    appium:deviceName=${DEVICE_NAME}
+    ...    appium:automationName=${AUTOMATION_NAME}
+    ...    appium:appPackage=${APP_PACKAGE}
+    ...    appium:appActivity=${APP_ACTIVITY}
+    ...    appium:noReset=${TRUE}
+
diff --git a/pyproject.toml b/pyproject.toml
index b8f9871..de28dd9 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -7,7 +7,7 @@ name = "robotframework-agent"
 version = "0.0.1"
 description = "AI-driven test automation agent for apps"
 readme = "readme.md"
-requires-python = ">=3.8"
+requires-python = ">=3.10"
 license = {text = "CC0-1.0"}
 authors = [
     {name = "Abdelkader HASSINE", email = "contact.abdelkaderhassine@gmail.com"},
@@ -20,11 +20,10 @@ classifiers = [
     "Topic :: Software Development :: Testing",
     "License :: CC0 1.0 Universal (CC0 1.0) Public Domain Dedication",
     "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3.8",
-    "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
     "Framework :: Robot Framework",
     "Framework :: Robot Framework :: Library",
 ]
@@ -87,14 +86,31 @@ Repository = "https://github.com/aidriventesting/agent"
 Issues = "https://github.com/aidriventesting/agent/issues"
 
 [tool.setuptools]
-packages = ["Agent", "Agent.ai", "Agent.ai.llm", "Agent.ai.vlm", "Agent.config", "Agent.platforms", "Agent.utilities", "Agent.utilities.imguploader"]
+packages = [
+    "Agent",
+    "Agent.ai",
+    "Agent.ai.llm",
+    "Agent.ai.vlm",
+    "Agent.ai.prompts",
+    "Agent.config",
+    "Agent.core",
+    "Agent.platforms",
+    "Agent.platforms.collectors",
+    "Agent.platforms.locators",
+    "Agent.tools",
+    "Agent.tools.mobile",
+    "Agent.tools.visual",
+    "Agent.tools.web",
+    "Agent.utilities",
+    "Agent.utilities.imguploader",
+]
 
 [tool.setuptools.package-data]
 Agent = ["config/*.json"]
 
 [tool.black]
 line-length = 100
-target-version = ['py38', 'py39', 'py310', 'py311', 'py312']
+target-version = ['py310', 'py311', 'py312', 'py313']
 
 [tool.pytest.ini_options]
 testpaths = ["tests/utest"]
diff --git a/readme.md b/readme.md
index 0ce787d..a7b11ca 100644
--- a/readme.md
+++ b/readme.md
@@ -10,7 +10,7 @@ Alpha — An evolving experiment, with varying levels of maturity across keyword
 
 ```robot
 *** Settings ***
-Library    Agent    llm_client=openai    llm_model=gpt-4o-mini
+Library    Agent    llm_client=openai    llm_model=gpt-4.1    platform_type=mobile    element_source=accessibility
 
 *** Test Cases ***
 Login
@@ -26,7 +26,7 @@ Login
 # Core
 pip install robotframework-agent
 
-# Web testing (+ Playwright)
+# Web testing (coming soon)
 pip install robotframework-agent[web]
 
 # Mobile testing (+ Appium)
@@ -36,6 +36,26 @@ pip install robotframework-agent[mobile]
 pip install robotframework-agent[dev]
 ```
 
+## LLM Providers
+
+Supports OpenAI (default), Anthropic Claude, and Google Gemini.
+
+```bash
+# With Anthropic/Claude
+pip install robotframework-agent[anthropic]
+
+# With Google Gemini
+pip install robotframework-agent[gemini]
+```
+
+```robot
+# Use Claude
+Library    Agent    llm_client=anthropic    llm_model=claude-sonnet-4
+
+# Use Gemini
+Library    Agent    llm_client=gemini    llm_model=gemini-2.0-flash
+```
+
 ## Keywords
 
 **Agent.Do** `<instruction>`
@@ -60,11 +80,12 @@ pip install robotframework-agent[dev]
 Instruction → LLM → UI Context → Tool Selection → Execution
 ```
 
+Experiments and design choices are informed by research on AI agents and UI perception:
+- Support Vision-based UI parsing using OmniParser for element detection
+- Set-of-Mark (SoM) technique for visual grounding
+- Multi-provider LLM support (OpenAI, Anthropic, Gemini)
 
-Design choices are informed by research on AI agents and UI perception.
-Ideas are iterated and evaluated in our evolving lab suite:
-[AgentArena](https://github.com/aidriventesting/AgentArena).
-
+Ideas are tested and refined in [AgentArena](https://github.com/aidriventesting/AgentArena), our experimental testing environment.
 
 ## Presented at RoboCon 2026 (Helsinki)
 
diff --git a/tests/_data/images/icons/BlaBlaCar.png b/tests/_data/images/icons/BlaBlaCar.png
deleted file mode 100644
index 01408e3..0000000
Binary files a/tests/_data/images/icons/BlaBlaCar.png and /dev/null differ
diff --git a/tests/_data/images/icons/Citygoo.png b/tests/_data/images/icons/Citygoo.png
deleted file mode 100644
index e707262..0000000
Binary files a/tests/_data/images/icons/Citygoo.png and /dev/null differ
diff --git a/tests/_data/images/icons/Cityscoot.png b/tests/_data/images/icons/Cityscoot.png
deleted file mode 100644
index a8125a8..0000000
Binary files a/tests/_data/images/icons/Cityscoot.png and /dev/null differ
diff --git a/tests/_data/images/icons/Comunauto.png b/tests/_data/images/icons/Comunauto.png
deleted file mode 100644
index ff0613f..0000000
Binary files a/tests/_data/images/icons/Comunauto.png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (1).png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (1).png
deleted file mode 100644
index 7f2588c..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (1).png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (2).png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (2).png
deleted file mode 100644
index bae52f0..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (2).png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (3).png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (3).png
deleted file mode 100644
index e3071f6..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (3).png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (4).png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (4).png
deleted file mode 100644
index 57d8c26..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (4).png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (5).png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (5).png
deleted file mode 100644
index 5330f58..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (5).png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (6).png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (6).png
deleted file mode 100644
index 437fed8..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0 (6).png and /dev/null differ
diff --git a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0.png b/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0.png
deleted file mode 100644
index 7d185e6..0000000
Binary files a/tests/_data/images/screenshots/screenshot-Google Pixel 5-11.0.png and /dev/null differ
diff --git a/tests/_data/xml/facebook-createaccount-screen.xml b/tests/_data/xml/facebook-createaccount-screen.xml
deleted file mode 100644
index f65886c..0000000
--- a/tests/_data/xml/facebook-createaccount-screen.xml
+++ /dev/null
@@ -1,44 +0,0 @@
-<?xml version='1.0' encoding='UTF-8' standalone='yes' ?>
-<hierarchy index="0" class="hierarchy" rotation="0" width="1080" height="2160">
-  <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-    <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-      <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="android:id/content" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,145][1080,2160]" displayed="true">
-        <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,145][1080,2160]" displayed="true">
-          <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,145][1080,288]" displayed="true">
-            <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,145][1080,288]" displayed="true">
-              <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,145][1080,288]" displayed="true">
-                <android.widget.ImageView index="0" package="com.facebook.katana" class="android.widget.ImageView" text="" content-desc="Retour" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,145][154,288]" displayed="true" />
-                <android.widget.FrameLayout index="1" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[154,145][1036,288]" displayed="true">
-                  <android.widget.TextView index="0" package="com.facebook.katana" class="android.widget.TextView" text="Créer un compte" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[154,145][525,288]" displayed="true" />
-                </android.widget.FrameLayout>
-                <android.widget.LinearLayout index="2" package="com.facebook.katana" class="android.widget.LinearLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[1036,145][1080,288]" displayed="true" />
-              </android.widget.LinearLayout>
-              <android.view.View index="1" package="com.facebook.katana" class="android.view.View" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,286][1080,288]" displayed="true" />
-            </android.widget.FrameLayout>
-          </android.widget.LinearLayout>
-          <android.widget.FrameLayout index="1" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,288][1080,2160]" displayed="true">
-            <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,288][1080,2160]" displayed="true">
-              <android.widget.ScrollView index="0" package="com.facebook.katana" class="android.widget.ScrollView" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,288][1080,2160]" displayed="true">
-                <android.widget.RelativeLayout index="0" package="com.facebook.katana" class="android.widget.RelativeLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,288][1080,2160]" displayed="true">
-                  <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,288][1080,2160]" displayed="true">
-                    <android.widget.ImageView index="0" package="com.facebook.katana" class="android.widget.ImageView" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,288][1025,925]" displayed="true" />
-                    <android.view.View index="1" package="com.facebook.katana" class="android.view.View" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,925][1025,1073]" displayed="true" />
-                    <android.widget.TextView index="2" package="com.facebook.katana" class="android.widget.TextView" text="Rejoignez Facebook" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,1106][1025,1173]" displayed="true" />
-                    <android.widget.TextView index="3" package="com.facebook.katana" class="android.widget.TextView" text="Nous vous aiderons à créer un compte en quelques étapes." resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,1206][1025,1313]" displayed="true" />
-                    <android.view.View index="4" package="com.facebook.katana" class="android.view.View" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,1313][1025,1462]" displayed="true" />
-                    <android.widget.Button index="5" package="com.facebook.katana" class="android.widget.Button" text="Suivant" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,1462][1025,1581]" displayed="true" />
-                    <android.view.View index="6" package="com.facebook.katana" class="android.view.View" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,1581][1025,2160]" displayed="true" />
-                    <android.widget.LinearLayout index="7" package="com.facebook.katana" class="android.widget.LinearLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[55,2177][1025,2296]" displayed="true">
-                      <android.widget.Button index="0" package="com.facebook.katana" class="android.widget.Button" text="Vous avez déjà un compte ?" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[262,2177][818,2296]" displayed="true" />
-                    </android.widget.LinearLayout>
-                  </android.widget.LinearLayout>
-                </android.widget.RelativeLayout>
-              </android.widget.ScrollView>
-            </android.widget.FrameLayout>
-          </android.widget.FrameLayout>
-        </android.widget.LinearLayout>
-      </android.widget.FrameLayout>
-    </android.widget.LinearLayout>
-    <android.view.View index="2" package="com.facebook.katana" class="android.view.View" text="" resource-id="android:id/navigationBarBackground" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,2296][1080,2340]" displayed="true" />
-  </android.widget.FrameLayout>
-</hierarchy>
\ No newline at end of file
diff --git a/tests/_data/xml/facebook-login-screen.xml b/tests/_data/xml/facebook-login-screen.xml
deleted file mode 100644
index 10c19e8..0000000
--- a/tests/_data/xml/facebook-login-screen.xml
+++ /dev/null
@@ -1,28 +0,0 @@
-<?xml version='1.0' encoding='UTF-8' standalone='yes' ?>
-<hierarchy index="0" class="hierarchy" rotation="0" width="1080" height="2160">
-  <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-    <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-      <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="android:id/content" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-        <android.widget.LinearLayout index="0" package="com.facebook.katana" class="android.widget.LinearLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-          <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-            <android.widget.FrameLayout index="0" package="com.facebook.katana" class="android.widget.FrameLayout" text="" resource-id="com.facebook.katana:id/(name removed)" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-              <android.view.ViewGroup index="0" package="com.facebook.katana" class="android.view.ViewGroup" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,0][1080,2160]" displayed="true">
-                <android.view.ViewGroup index="0" package="com.facebook.katana" class="android.view.ViewGroup" text="" checkable="false" checked="false" clickable="true" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[314,637][439,690]" displayed="true" />
-                <android.view.ViewGroup index="1" package="com.facebook.katana" class="android.view.ViewGroup" text="" checkable="false" checked="false" clickable="true" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[486,637][591,690]" displayed="true" />
-                <android.view.ViewGroup index="2" package="com.facebook.katana" class="android.view.ViewGroup" text="" content-desc="Autres langues..." checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[638,637][766,690]" displayed="true" />
-                <android.widget.EditText index="3" package="com.facebook.katana" class="android.widget.EditText" text="Téléphone ou e-mail" content-desc="Nom d’utilisateur" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="true" password="false" scrollable="false" selected="false" bounds="[88,813][992,951]" displayed="true" />
-                <android.widget.EditText index="4" package="com.facebook.katana" class="android.widget.EditText" text="Mot de passe" content-desc="Mot de passe" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="true" password="true" scrollable="false" selected="false" bounds="[88,973][992,1111]" displayed="true" />
-                <android.view.ViewGroup index="5" package="com.facebook.katana" class="android.view.ViewGroup" text="" content-desc="Se connecter" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[88,1155][992,1265]" displayed="true" />
-                <android.view.ViewGroup index="6" package="com.facebook.katana" class="android.view.ViewGroup" text="" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,1265][1080,1415]" displayed="true">
-                  <android.view.ViewGroup index="0" package="com.facebook.katana" class="android.view.ViewGroup" text="" content-desc="Mot de passe oublié ?" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[314,1298][766,1409]" displayed="true" />
-                </android.view.ViewGroup>
-                <android.view.ViewGroup index="7" package="com.facebook.katana" class="android.view.ViewGroup" text="" content-desc="Créer un compte Facebook" checkable="false" checked="false" clickable="true" enabled="true" focusable="true" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[229,1695][852,1794]" displayed="true" />
-              </android.view.ViewGroup>
-            </android.widget.FrameLayout>
-          </android.widget.FrameLayout>
-        </android.widget.LinearLayout>
-      </android.widget.FrameLayout>
-    </android.widget.LinearLayout>
-    <android.view.View index="1" package="com.facebook.katana" class="android.view.View" text="" resource-id="android:id/navigationBarBackground" checkable="false" checked="false" clickable="false" enabled="true" focusable="false" focused="false" long-clickable="false" password="false" scrollable="false" selected="false" bounds="[0,2296][1080,2340]" displayed="true" />
-  </android.widget.FrameLayout>
-</hierarchy>
\ No newline at end of file
diff --git a/tests/atest/e2e/atest_mobile_b_contactform.robot b/tests/atest/acceptance/mobile/flow_contact_form.robot
similarity index 95%
rename from tests/atest/e2e/atest_mobile_b_contactform.robot
rename to tests/atest/acceptance/mobile/flow_contact_form.robot
index 2005ddb..04a8a3e 100644
--- a/tests/atest/e2e/atest_mobile_b_contactform.robot
+++ b/tests/atest/acceptance/mobile/flow_contact_form.robot
@@ -2,7 +2,7 @@
 Documentation    E2E test for Contact Us form - Happy path
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test Contact Us Form Submit
diff --git a/tests/atest/e2e/atest_mobile_i_hybrid.robot b/tests/atest/acceptance/mobile/flow_hybrid.robot
similarity index 93%
rename from tests/atest/e2e/atest_mobile_i_hybrid.robot
rename to tests/atest/acceptance/mobile/flow_hybrid.robot
index 30e0fcc..ddc430f 100644
--- a/tests/atest/e2e/atest_mobile_i_hybrid.robot
+++ b/tests/atest/acceptance/mobile/flow_hybrid.robot
@@ -2,7 +2,7 @@
 Documentation    E2E test for Hybrid Activity - Happy path
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test Hybrid Activity
diff --git a/tests/atest/e2e/atest_mobile_e_login.robot b/tests/atest/acceptance/mobile/flow_login.robot
similarity index 96%
rename from tests/atest/e2e/atest_mobile_e_login.robot
rename to tests/atest/acceptance/mobile/flow_login.robot
index 6fba264..d7564de 100644
--- a/tests/atest/e2e/atest_mobile_e_login.robot
+++ b/tests/atest/acceptance/mobile/flow_login.robot
@@ -2,7 +2,7 @@
 Documentation    E2E test for login functionality - Happy path
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test Login Success
diff --git a/tests/atest/e2e/atest_mobile_d_navigation.robot b/tests/atest/acceptance/mobile/flow_navigation.robot
similarity index 94%
rename from tests/atest/e2e/atest_mobile_d_navigation.robot
rename to tests/atest/acceptance/mobile/flow_navigation.robot
index 2c06ad9..195433c 100644
--- a/tests/atest/e2e/atest_mobile_d_navigation.robot
+++ b/tests/atest/acceptance/mobile/flow_navigation.robot
@@ -2,7 +2,7 @@
 Documentation    E2E test for tab navigation - Happy path
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test Tab Navigation
diff --git a/tests/atest/acceptance/mobile/idfm.robot b/tests/atest/acceptance/mobile/idfm.robot
new file mode 100644
index 0000000..3cf7d5c
--- /dev/null
+++ b/tests/atest/acceptance/mobile/idfm.robot
@@ -0,0 +1,19 @@
+*** Settings ***
+Documentation    E2E test for IDFM - Happy path
+Library    AppiumLibrary
+Library    Agent
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
+
+
+*** Test Cases ***
+Test IDFM
+    [Documentation]    Happy path: Navigate between HOME, SPORT, and MOVIE tabs
+    [Tags]    mobile    navigation
+    Open Application With Config
+    Sleep    1s
+    
+    Agent.do    instruction=click on the TAB ACTIVITY button
+    Sleep    1s
+    
+    Agent.check    instruction=verify that the HOME tab is selected and Home fragment text is displayed
+    Sleep    1s
\ No newline at end of file
diff --git a/tests/atest/e2e/atest_web_arena_benchmark.robot b/tests/atest/acceptance/web/arena_benchmark.robot
similarity index 67%
rename from tests/atest/e2e/atest_web_arena_benchmark.robot
rename to tests/atest/acceptance/web/arena_benchmark.robot
index fd891c3..45eb386 100644
--- a/tests/atest/e2e/atest_web_arena_benchmark.robot
+++ b/tests/atest/acceptance/web/arena_benchmark.robot
@@ -1,5 +1,5 @@
 *** Settings ***
-Documentation    E2E test for Agent Arena benchmark - Tests different click_mode and input_mode combinations
+Documentation    E2E test for Agent Arena benchmark - Tests different element_source and selection_mode combinations
 ...              Website: https://agent-arena-eta.vercel.app/webarena/c0/
 ...    
 ...              Run all combinations:
@@ -8,35 +8,29 @@ Documentation    E2E test for Agent Arena benchmark - Tests different click_mode
 ...              Run specific combination:
 ...                  robot --include som tests/atest/e2e/atest_web_arena_benchmark.robot
 ...                  robot --include text tests/atest/e2e/atest_web_arena_benchmark.robot
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 Test Template    Run Arena Test
 
 *** Variables ***
 ${ARENA_URL}    https://agent-arena-eta.vercel.app/webarena/c0/feed.html
 ${DELAY}        2
 
-*** Test Cases ***                  CLICK_MODE    INPUT_MODE
-Text + XML (default)                xml           text
-    [Tags]    text    xml
+*** Test Cases ***                  ELEMENT_SOURCE    LLM_INPUT_FORMAT
+Text + Tree (default)               dom              text
+    [Tags]    text    dom
 
-SoM + XML                           xml           som
-    [Tags]    som    xml
+SoM + Tree                          dom              som
+    [Tags]    som    dom
 
-Visual Only                         visual        som
+Visual Only                         visual            visual
     [Tags]    visual
 
 *** Keywords ***
 Run Arena Test
     [Documentation]    Execute the arena benchmark test with specified modes
-    [Arguments]    ${click_mode}    ${input_mode}
+    [Arguments]    ${element_source}    ${llm_input_format}
     
-    Log    Testing: click_mode=${click_mode}, input_mode=${input_mode}    console=True
-    
-    ${agent}=    Get Library Instance    Agent
-    Call Method    ${agent.engine}    set_click_mode    ${click_mode}
-    Call Method    ${agent.engine}    set_input_mode    ${input_mode}
-    
-    Open Website    ${ARENA_URL}
+    Open Website    ${ARENA_URL}    ${element_source}    ${llm_input_format}
     Sleep    ${DELAY}s
     
     # Test 1: Navigate via sidebar
diff --git a/tests/atest/e2e/atest_web_z_complete_form_scenario.robot b/tests/atest/acceptance/web/flow_form.robot
similarity index 98%
rename from tests/atest/e2e/atest_web_z_complete_form_scenario.robot
rename to tests/atest/acceptance/web/flow_form.robot
index 53832d2..831cfc1 100644
--- a/tests/atest/e2e/atest_web_z_complete_form_scenario.robot
+++ b/tests/atest/acceptance/web/flow_form.robot
@@ -2,7 +2,7 @@
 Documentation    Complete E2E scenario on a real website: Wikipedia article exploration
 ...              This test simulates a real user journey: search, navigate, read, and explore Wikipedia
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Variables ***
 ${WIKIPEDIA_URL}    https://en.wikipedia.org
diff --git a/tests/atest/e2e/atest_web_vision_wikipedia.robot b/tests/atest/acceptance/web/flow_wikipedia.robot
similarity index 77%
rename from tests/atest/e2e/atest_web_vision_wikipedia.robot
rename to tests/atest/acceptance/web/flow_wikipedia.robot
index 5dcbbd5..8e5272a 100644
--- a/tests/atest/e2e/atest_web_vision_wikipedia.robot
+++ b/tests/atest/acceptance/web/flow_wikipedia.robot
@@ -1,27 +1,28 @@
 *** Settings ***
 Documentation    E2E tests for web vision mode on real-world websites
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
-Suite Setup    Open Website
-Suite Teardown    Close Browser
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
+
 
 *** Test Cases ***
 Test Click Wikipedia Link
     [Documentation]    Test clicking a link on Wikipedia homepage using vision mode
     [Tags]    vision    web    click
     
-    Go To    https://www.wikipedia.org/
+    Open Website    https://www.wikipedia.org/
     Sleep    2s
     
-    Agent.Do    click on English link
+    Agent.Do    click on English language 
     Sleep    1s
-    
-    Get Url    should contain    en.wikipedia.org
+
+    agent.do    click on the search input fields
+
+    agent.do    type "Robot Framework" in the search input fields
 
 Test Search On DuckDuckGo
     [Documentation]    Test searching on DuckDuckGo using vision mode
     [Tags]    vision    web    input    search
     
-    Go To    https://duckduckgo.com/
+    Open Website    https://duckduckgo.com/
     Sleep    2s
     
     Agent.Do    type 'robot framework' in the search box
@@ -36,7 +37,7 @@ Test Navigate GitHub
     [Documentation]    Test navigating GitHub homepage using vision mode
     [Tags]    vision    web    navigation
     
-    Go To    https://github.com/
+    Open Website    https://github.com/
     Sleep    2s
     
     Agent.Do    click on Sign in button
@@ -48,7 +49,7 @@ Test Click Google Search Button
     [Documentation]    Test clicking Google search button using vision mode
     [Tags]    vision    web    click
     
-    Go To    https://www.google.com/
+    Open Website    https://www.google.com/
     Sleep    2s
     
     # Accept cookies if present (might fail, that's ok)
diff --git a/tests/atest/e2e/atest_web_vision_youtube.robot b/tests/atest/acceptance/web/flow_youtube.robot
similarity index 80%
rename from tests/atest/e2e/atest_web_vision_youtube.robot
rename to tests/atest/acceptance/web/flow_youtube.robot
index 65109af..b115564 100644
--- a/tests/atest/e2e/atest_web_vision_youtube.robot
+++ b/tests/atest/acceptance/web/flow_youtube.robot
@@ -1,7 +1,6 @@
 *** Settings ***
 Documentation    Real-world scenario tests using vision mode on popular websites
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
-Suite Setup    Open Website
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 Suite Teardown    Close Browser
 
 *** Test Cases ***
@@ -9,40 +8,37 @@ Test YouTube Search And Interact
     [Documentation]    Search on YouTube, play video, read comment and like it
     [Tags]    vision    web    youtube    real-scenario
     
-    Go To    https://www.youtube.com/
-    Sleep    3s
+    Open Website    https://www.youtube.com/
+    # Sleep    3s
     
     # Accept cookies if present
-    Run Keyword And Ignore Error    Agent.Do    click on Accept all button
-    Sleep    2s
-    
-    # Search for a video
-    Agent.Do    click on search box
-    Sleep    1s
+    Agent.Do    click on Accept coockies button to disappear popup
+    # Sleep    2s        
     
     Agent.Do    type 'robot framework tutorial' in search box
-    Sleep    1s
+    # Sleep    1s
+
+    Sleep    2
     
     Agent.Do    press Enter key
-    Sleep    3s
+    # Sleep    3s
     
     # Click on first video
-    Agent.Do    click on the first video thumbnail
-    Sleep    5s
+    Agent.Do    click on the video of of robot framework introduction by Pekka Klärck which appears on the search results of youtube 
+    # Sleep    5s
     
     # Scroll down to comments
     Agent.Do    scroll down
-    Sleep    2s
-    Agent.Do    scroll down
-    Sleep    2s
+    # Sleep    2s
+
     
     # Read first comment (using check to verify it's visible)
     Agent.check    instruction=verify that comments section is visible
-    Sleep    1s
+    # Sleep    1s
     
     # Like the first comment
     Agent.Do    click on like button of the first comment
-    Sleep    1s
+    # Sleep    1s
     
     # Verify the like was registered
     Agent.check    instruction=verify that the like button is highlighted
@@ -51,7 +47,7 @@ Test Amazon Product Search
     [Documentation]    Search for a product on Amazon and add to cart
     [Tags]    vision    web    amazon    real-scenario
     
-    Go To    https://www.amazon.com/
+    Open Website    https://www.amazon.com/
     Sleep    3s
     
     # Search for product
@@ -75,7 +71,7 @@ Test Twitter Browse
     [Documentation]    Browse Twitter/X homepage
     [Tags]    vision    web    twitter    real-scenario
     
-    Go To    https://twitter.com/
+    Open Website    https://twitter.com/
     Sleep    3s
     
     # Verify homepage elements
@@ -89,7 +85,7 @@ Test Reddit Browse Subreddit
     [Documentation]    Browse a subreddit on Reddit
     [Tags]    vision    web    reddit    real-scenario
     
-    Go To    https://www.reddit.com/r/robotframework/
+    Open Website    https://www.reddit.com/r/robotframework/
     Sleep    3s
     
     # Scroll to see posts
@@ -110,7 +106,7 @@ Test LinkedIn Profile Navigation
     [Documentation]    Navigate LinkedIn homepage
     [Tags]    vision    web    linkedin    real-scenario
     
-    Go To    https://www.linkedin.com/
+    Open Website    https://www.linkedin.com/
     Sleep    3s
     
     # Verify sign in is available
diff --git a/tests/atest/browsers/config.robot b/tests/atest/browsers/config.robot
deleted file mode 100644
index e122944..0000000
--- a/tests/atest/browsers/config.robot
+++ /dev/null
@@ -1,14 +0,0 @@
-*** Settings ***
-Documentation    Configuration centralisée pour les tests web
-Variables    ${EXECDIR}/tests/atest/browsers/settings.yaml
-Library    Browser
-Library    Agent    llm_client=openai    llm_model=gpt-4o-mini
-*** Keywords ***
-Open Website
-    [Documentation]    Ouvre le navigateur selon la configuration. Optionnellement navigue vers une URL.
-    ...                Args: url (optionnel) - URL vers laquelle naviguer. Si vide, ouvre seulement le navigateur.
-    [Arguments]    ${url}=${EMPTY}
-    New Browser    ${browser_type}    headless=${headless_mode}
-    Log    Browser opened - Type: ${browser_type}, Headless: ${headless_mode}
-    New Page    ${url}
-    
\ No newline at end of file
diff --git a/tests/atest/browsers/settings.yaml b/tests/atest/browsers/settings.yaml
deleted file mode 100644
index ee6479a..0000000
--- a/tests/atest/browsers/settings.yaml
+++ /dev/null
@@ -1,27 +0,0 @@
-# ============================================================================
-# BROWSER CONFIGURATION
-# ============================================================================
-# See README.md in this folder for detailed documentation
-#
-# Quick Start:
-# 1. Configure browser type and headless mode below
-# 2. Run: robot tests/atest/e2e/your_test.robot
-#
-# ============================================================================
-# Global Settings
-# ============================================================================
-browser_type: chromium  # Options: chromium | firefox | webkit
-headless_mode: True       # Options: True | False
-
-# Common Test URLs
-form_url: https://www.selenium.dev/selenium/web/web-form.html
-long_page_url: https://robotframework.org/
-home_url: https://www.selenium.dev
-
-# Wait strategy
-delay: 0.3
-veryShort_wait: 1
-short_wait: 2
-normal_wait: 3
-long_wait: 5
-
diff --git a/tests/atest/compounent/BBoxVisualizer.py b/tests/atest/compounent/BBoxVisualizer.py
deleted file mode 100644
index 5f17cac..0000000
--- a/tests/atest/compounent/BBoxVisualizer.py
+++ /dev/null
@@ -1,413 +0,0 @@
-"""
-BBoxVisualizer - Robot Framework Library for visualizing bounding boxes.
-
-This library provides keywords to draw bounding boxes on screenshots,
-making it easy to visualize which elements were detected by UI collectors.
-
-Usage:
-    Library    tests/atest/compounent/BBoxVisualizer.py
-    
-    ${annotated}=    Draw And Log Elements    ${elements}    screenshot.png
-"""
-
-import os
-import base64
-from datetime import datetime
-from typing import List, Dict, Any, Optional
-from PIL import Image, ImageDraw, ImageFont
-from robot.api import logger
-from robot.libraries.BuiltIn import BuiltIn
-
-
-class BBoxVisualizer:
-    """
-    Robot Framework library for visualizing bounding boxes on screenshots.
-    
-    This library helps debug and visualize UI element detection by drawing
-    rectangles around detected elements with numbers and labels.
-    """
-    
-    ROBOT_LIBRARY_SCOPE = 'GLOBAL'
-    ROBOT_LIBRARY_VERSION = '1.0.0'
-    
-    def __init__(self):
-        """Initialize the BBoxVisualizer library."""
-        self.default_color = 'red'
-        self.default_width = 3
-        self.default_text_color = 'white'
-        self.default_label_color = 'blue'
-        self.builtin = BuiltIn()
-    
-    def _resolve_path(self, path: str) -> str:
-        """
-        Resolve a path that might be relative to Robot Framework directories.
-        
-        Tries in order:
-        1. Path as-is (if absolute or exists)
-        2. Relative to OUTPUT_DIR
-        3. Relative to EXECDIR
-        4. Relative to current working directory
-        
-        Args:
-            path: Path to resolve
-            
-        Returns:
-            Resolved absolute path
-            
-        Raises:
-            FileNotFoundError: If path cannot be resolved
-        """
-        # If already absolute and exists, return it
-        if os.path.isabs(path) and os.path.exists(path):
-            return path
-        
-        # If relative path exists from current location, return it
-        if os.path.exists(path):
-            return os.path.abspath(path)
-        
-        # Try to get Robot Framework variables
-        try:
-            output_dir = self.builtin.get_variable_value("${OUTPUT DIR}")
-            if output_dir:
-                resolved = os.path.join(output_dir, path)
-                if os.path.exists(resolved):
-                    logger.debug(f"Resolved path using OUTPUT_DIR: {resolved}")
-                    return resolved
-        except:
-            pass
-        
-        try:
-            execdir = self.builtin.get_variable_value("${EXECDIR}")
-            if execdir:
-                resolved = os.path.join(execdir, path)
-                if os.path.exists(resolved):
-                    logger.debug(f"Resolved path using EXECDIR: {resolved}")
-                    return resolved
-        except:
-            pass
-        
-        # Try current working directory
-        resolved = os.path.abspath(path)
-        if os.path.exists(resolved):
-            logger.debug(f"Resolved path using CWD: {resolved}")
-            return resolved
-        
-        # If still not found, raise error with helpful message
-        logger.error(f"Could not resolve path: {path}")
-        logger.error(f"Tried:")
-        logger.error(f"  - As-is: {path}")
-        logger.error(f"  - CWD: {os.path.abspath(path)}")
-        try:
-            output_dir = self.builtin.get_variable_value("${OUTPUT DIR}")
-            if output_dir:
-                logger.error(f"  - OUTPUT_DIR: {os.path.join(output_dir, path)}")
-        except:
-            pass
-        
-        raise FileNotFoundError(f"Could not find file: {path}")
-    
-    def draw_bbox_on_screenshot(
-        self, 
-        screenshot_path: str, 
-        elements: List[Dict[str, Any]], 
-        output_path: Optional[str] = None,
-        color: str = 'red',
-        width: int = 3,
-        show_numbers: bool = True,
-        show_labels: bool = True
-    ) -> str:
-        """
-        Draw bounding boxes on a screenshot.
-        
-        Args:
-            screenshot_path: Path to the original image
-            elements: List of elements with bbox information
-            output_path: Output path (optional, auto-generated if None)
-            color: Color of the rectangles (default: red)
-            width: Width of the rectangle lines (default: 3)
-            show_numbers: Whether to show element numbers (default: True)
-            show_labels: Whether to show element labels (default: True)
-        
-        Returns:
-            Path to the annotated image file
-        
-        Example:
-            ${annotated}=    Draw BBox On Screenshot    screenshot.png    ${elements}
-        """
-        logger.info(f"Drawing bounding boxes on {screenshot_path}")
-        logger.info(f"Number of elements to annotate: {len(elements)}")
-        
-        # Resolve path (handle Robot Framework relative paths)
-        screenshot_path = self._resolve_path(screenshot_path)
-        logger.debug(f"Resolved screenshot path: {screenshot_path}")
-        
-        # Load image
-        try:
-            img = Image.open(screenshot_path)
-        except Exception as e:
-            logger.error(f"Failed to open screenshot: {e}")
-            raise
-        
-        draw = ImageDraw.Draw(img)
-        
-        # Try to load a font, fallback to default if not available
-        try:
-            font_size = 16
-            font = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", font_size)
-            font_large = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", 20)
-        except:
-            try:
-                font = ImageFont.truetype("/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf", 16)
-                font_large = ImageFont.truetype("/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf", 20)
-            except:
-                font = ImageFont.load_default()
-                font_large = ImageFont.load_default()
-        
-        # Draw each bounding box
-        annotated_count = 0
-        for i, elem in enumerate(elements, 1):
-            bbox = elem.get('bbox', {})
-            if not bbox or not all(k in bbox for k in ['x', 'y', 'width', 'height']):
-                logger.warn(f"Element {i} missing valid bbox: {bbox}")
-                continue
-            
-            x = bbox['x']
-            y = bbox['y']
-            w = bbox['width']
-            h = bbox['height']
-            
-            # Draw rectangle
-            draw.rectangle(
-                [x, y, x + w, y + h],
-                outline=color,
-                width=width
-            )
-            
-            # Draw number badge (top-left corner)
-            if show_numbers:
-                number_text = str(i)
-                
-                # Calculate text size
-                bbox_text = draw.textbbox((0, 0), number_text, font=font_large)
-                text_width = bbox_text[2] - bbox_text[0]
-                text_height = bbox_text[3] - bbox_text[1]
-                
-                # Position for number (above the element)
-                number_x = x
-                number_y = max(0, y - text_height - 8)
-                
-                # Draw white background for number
-                padding = 4
-                draw.rectangle(
-                    [number_x - padding, number_y - padding,
-                     number_x + text_width + padding, number_y + text_height + padding],
-                    fill='white',
-                    outline=color,
-                    width=2
-                )
-                
-                # Draw number
-                draw.text((number_x, number_y), number_text, fill=color, font=font_large)
-            
-            # Draw label (inside the box, top-left)
-            if show_labels:
-                label = elem.get('text') or elem.get('content_desc') or elem.get('name', '')
-                if label:
-                    # Truncate long labels
-                    label = label[:30]
-                    
-                    # Draw semi-transparent background for label
-                    label_x = x + 5
-                    label_y = y + 5
-                    
-                    bbox_label = draw.textbbox((0, 0), label, font=font)
-                    label_width = bbox_label[2] - bbox_label[0]
-                    label_height = bbox_label[3] - bbox_label[1]
-                    
-                    # Only draw if it fits inside the box
-                    if label_y + label_height < y + h:
-                        draw.rectangle(
-                            [label_x - 2, label_y - 2,
-                             label_x + label_width + 2, label_y + label_height + 2],
-                            fill='white',
-                            outline=None
-                        )
-                        draw.text((label_x, label_y), label, fill=self.default_label_color, font=font)
-            
-            annotated_count += 1
-        
-        # Generate output path if not provided
-        if not output_path:
-            base, ext = os.path.splitext(screenshot_path)
-            output_path = f"{base}_annotated{ext}"
-        
-        # Ensure output directory exists
-        output_dir = os.path.dirname(output_path)
-        if output_dir and not os.path.exists(output_dir):
-            os.makedirs(output_dir, exist_ok=True)
-            logger.debug(f"Created output directory: {output_dir}")
-        
-        # Save annotated image
-        img.save(output_path)
-        logger.info(f"Annotated {annotated_count} elements, saved to: {output_path}")
-        
-        return output_path
-    
-    def annotate_elements(
-        self, 
-        elements: List[Dict[str, Any]], 
-        screenshot_path: str, 
-        output_dir: str = "browser/annotated"
-    ) -> str:
-        """
-        Annotate elements on a screenshot with timestamp.
-        
-        Args:
-            elements: List of elements with bbox
-            screenshot_path: Path to original screenshot
-            output_dir: Directory for output (default: browser/annotated)
-        
-        Returns:
-            Path to the annotated image
-        
-        Example:
-            ${annotated}=    Annotate Elements    ${elements}    screenshot.png
-        """
-        # Create output directory
-        os.makedirs(output_dir, exist_ok=True)
-        
-        # Generate timestamped filename
-        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-        base_name = os.path.basename(screenshot_path)
-        name, ext = os.path.splitext(base_name)
-        output_path = os.path.join(output_dir, f"{name}_annotated_{timestamp}{ext}")
-        
-        return self.draw_bbox_on_screenshot(screenshot_path, elements, output_path)
-    
-    def draw_and_log_elements(
-        self, 
-        elements: List[Dict[str, Any]], 
-        screenshot_path: str,
-        width: int = 800
-    ) -> str:
-        """
-        Draw bounding boxes AND embed the annotated image in Robot Framework log.
-        
-        This is the most convenient method for Robot Framework tests as it
-        both creates the annotated image and displays it in the HTML log.
-        
-        Args:
-            elements: List of elements with bbox
-            screenshot_path: Path to original screenshot
-            width: Width for embedded image in log (default: 800)
-        
-        Returns:
-            Path to the annotated image
-        
-        Example:
-            ${annotated}=    Draw And Log Elements    ${elements}    screenshot.png
-        """
-        # Annotate the screenshot
-        annotated_path = self.annotate_elements(elements, screenshot_path)
-        
-        # Embed in Robot Framework log
-        try:
-            with open(annotated_path, 'rb') as f:
-                img_data = base64.b64encode(f.read()).decode()
-            
-            msg = f'</td></tr><tr><td colspan="3"><img src="data:image/png;base64,{img_data}" width="{width}"></td></tr>'
-            logger.info(msg, html=True, also_console=False)
-            logger.info(f"✅ Annotated screenshot embedded in log: {annotated_path}")
-        except Exception as e:
-            logger.warn(f"Failed to embed image in log: {e}")
-        
-        logger.info(f"Elements annotated:")
-        for i, element in enumerate(elements):
-            logger.info(f"Element {i}: {element}")
-        return annotated_path
-    
-    def create_comparison_image(
-        self,
-        screenshot_path: str,
-        elements_list1: List[Dict[str, Any]],
-        elements_list2: List[Dict[str, Any]],
-        label1: str = "Collector 1",
-        label2: str = "Collector 2",
-        output_path: Optional[str] = None
-    ) -> str:
-        """
-        Create a side-by-side comparison of two collectors.
-        
-        Args:
-            screenshot_path: Path to original screenshot
-            elements_list1: Elements from first collector
-            elements_list2: Elements from second collector
-            label1: Label for first collector
-            label2: Label for second collector
-            output_path: Output path (optional)
-        
-        Returns:
-            Path to comparison image
-        
-        Example:
-            ${comparison}=    Create Comparison Image    screenshot.png    ${js_elements}    ${ax_elements}    JSQuery    AXTree
-        """
-        # Annotate with first collector (red)
-        temp1 = self.draw_bbox_on_screenshot(
-            screenshot_path, 
-            elements_list1, 
-            screenshot_path.replace('.png', '_temp1.png'),
-            color='red'
-        )
-        
-        # Annotate with second collector (blue)
-        temp2 = self.draw_bbox_on_screenshot(
-            screenshot_path,
-            elements_list2,
-            screenshot_path.replace('.png', '_temp2.png'),
-            color='blue'
-        )
-        
-        # Load both images
-        img1 = Image.open(temp1)
-        img2 = Image.open(temp2)
-        
-        # Create side-by-side image
-        width = img1.width + img2.width
-        height = max(img1.height, img2.height) + 40  # +40 for labels
-        
-        combined = Image.new('RGB', (width, height), 'white')
-        
-        # Paste images
-        combined.paste(img1, (0, 40))
-        combined.paste(img2, (img1.width, 40))
-        
-        # Add labels
-        draw = ImageDraw.Draw(combined)
-        try:
-            font = ImageFont.truetype("/System/Library/Fonts/Helvetica.ttc", 24)
-        except:
-            font = ImageFont.load_default()
-        
-        draw.text((10, 10), f"{label1} ({len(elements_list1)} elements)", fill='red', font=font)
-        draw.text((img1.width + 10, 10), f"{label2} ({len(elements_list2)} elements)", fill='blue', font=font)
-        
-        # Save
-        if not output_path:
-            output_path = screenshot_path.replace('.png', '_comparison.png')
-        
-        combined.save(output_path)
-        
-        # Cleanup temp files
-        os.remove(temp1)
-        os.remove(temp2)
-        
-        logger.info(f"Comparison image saved: {output_path}")
-        return output_path
-
-
-if __name__ == "__main__":
-    # Quick test
-    print("BBoxVisualizer library loaded")
-    print("Version:", BBoxVisualizer.ROBOT_LIBRARY_VERSION)
-
diff --git a/tests/atest/compounent/README.md b/tests/atest/compounent/README.md
deleted file mode 100644
index 36ff8b1..0000000
--- a/tests/atest/compounent/README.md
+++ /dev/null
@@ -1,101 +0,0 @@
-# UI Collector Tests
-
-This directory contains Robot Framework tests for comparing different UI element collection strategies.
-
-## Available Collectors
-
-- **js_query**: JavaScript querySelector-based approach (original/default)
-- **axtree**: Accessibility Tree-based approach
-
-## Test Files
-
-### test_js_query.robot
-Tests the JSQuery collector in isolation on various websites.
-
-```bash
-robot tests/atest/collectors/test_js_query.robot
-```
-
-### test_axtree.robot
-Tests the AXTree collector in isolation on various websites.
-
-```bash
-robot tests/atest/collectors/test_axtree.robot
-```
-
-### compare_all.robot
-Compares both collectors side-by-side on the same pages to analyze differences.
-
-```bash
-robot tests/atest/collectors/compare_all.robot
-```
-
-## Test Sites
-
-The tests use the following websites:
-- **https://the-internet.herokuapp.com/login** - Simple login form
-- **https://www.saucedemo.com** - Demo e-commerce login
-- **https://react-shopping-cart-67954.firebaseapp.com/** - React SPA
-
-## Running Tests
-
-### Run all collector tests
-```bash
-robot tests/atest/collectors/
-```
-
-### Run with specific tags
-```bash
-# Only JSQuery tests
-robot --include js_query tests/atest/collectors/
-
-# Only AXTree tests
-robot --include axtree tests/atest/collectors/
-
-# Only comparison tests
-robot --include compare tests/atest/collectors/
-
-# Only login form tests
-robot --include login tests/atest/collectors/
-```
-
-## Expected Output
-
-The tests will log:
-- Number of elements found by each collector
-- Sample elements with their attributes
-- Comparison statistics (differences, element type counts)
-- Detailed element dictionaries for debugging
-
-## Adding New Collectors
-
-To add a new collector strategy:
-
-1. Create a new class in `Agent/platforms/collectors/` that inherits from `BaseUICollector`
-2. Register it in `collector_factory.py`
-3. Create a new test file `test_<name>.robot`
-4. Update `compare_all.robot` to include the new collector
-
-Example:
-```python
-# Agent/platforms/collectors/my_new_collector.py
-from Agent.platforms.collectors.base_collector import BaseUICollector
-
-class MyNewCollector(BaseUICollector):
-    def get_name(self) -> str:
-        return "my_new"
-    
-    def collect_elements(self, max_items: int = 150):
-        # Your implementation
-        pass
-
-# In collector_factory.py
-from Agent.platforms.collectors.my_new_collector import MyNewCollector
-CollectorRegistry.register("my_new", MyNewCollector)
-```
-
-Then use it:
-```robot
-Library    Agent.platforms._webconnector.WebConnectorRF    my_new    AS    Connector
-```
-
diff --git a/tests/atest/compounent/compare_all.robot b/tests/atest/compounent/compare_all.robot
deleted file mode 100644
index de96d59..0000000
--- a/tests/atest/compounent/compare_all.robot
+++ /dev/null
@@ -1,133 +0,0 @@
-*** Settings ***
-Documentation    Compare different collector strategies side-by-side
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    js_query    AS    JSQueryConnector
-Library          Agent.platforms._webconnector.WebConnectorRF    axtree    AS    AXTreeConnector
-Library          Collections
-
-
-*** Test Cases ***
-Compare Collectors On Simple Login Form
-    [Documentation]    Compare JSQuery and AXTree collectors on the same login form
-    [Tags]    collector    compare    login
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    # Collect with JSQuery
-    ${js_elements}=    JSQueryConnector.Collect UI Candidates
-    Log To Console    \n=== COMPARISON: Simple Login Form ===
-    Log To Console    JSQuery found: ${js_elements.__len__()} elements
-    
-    # Collect with AXTree
-    ${ax_elements}=    AXTreeConnector.Collect UI Candidates
-    Log To Console    AXTree found: ${ax_elements.__len__()} elements
-    
-    # Compare counts
-    ${diff}=    Evaluate    ${js_elements.__len__()} - ${ax_elements.__len__()}
-    Log To Console    Difference: ${diff} elements
-    
-    # Log some details
-    Log To Console    \n=== JSQuery First 5 Elements ===
-    Log To Console    ${js_elements[:5]}
-    
-    Log To Console    \n=== AXTree First 5 Elements ===
-    Log To Console    ${ax_elements[:5]}
-    
-    Should Not Be Empty    ${js_elements}
-    Should Not Be Empty    ${ax_elements}
-    
-    [Teardown]    Close Browser
-
-
-Compare Collectors On SauceDemo
-    [Documentation]    Compare collectors on SauceDemo login page
-    [Tags]    collector    compare    saucedemo
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://www.saucedemo.com
-    
-    # Collect with JSQuery
-    ${js_elements}=    JSQueryConnector.Collect UI Candidates
-    Log To Console    \n=== COMPARISON: SauceDemo ===
-    Log To Console    JSQuery found: ${js_elements.__len__()} elements
-    
-    # Collect with AXTree
-    ${ax_elements}=    AXTreeConnector.Collect UI Candidates
-    Log To Console    AXTree found: ${ax_elements.__len__()} elements
-    
-    # Compare counts
-    ${diff}=    Evaluate    ${js_elements.__len__()} - ${ax_elements.__len__()}
-    Log To Console    Difference: ${diff} elements
-    
-    # Analyze element types
-    ${js_inputs}=    Count Element Type    ${js_elements}    input
-    ${ax_inputs}=    Count Element Type    ${ax_elements}    input
-    Log To Console    \nInput fields - JSQuery: ${js_inputs}, AXTree: ${ax_inputs}
-    
-    ${js_buttons}=    Count Element Type    ${js_elements}    button
-    ${ax_buttons}=    Count Element Type    ${ax_elements}    button
-    Log To Console    Buttons - JSQuery: ${js_buttons}, AXTree: ${ax_buttons}
-    
-    ${js_links}=    Count Element Type    ${js_elements}    a
-    ${ax_links}=    Count Element Type    ${ax_elements}    a
-    Log To Console    Links - JSQuery: ${js_links}, AXTree: ${ax_links}
-    
-    [Teardown]    Close Browser
-
-
-Compare Collectors On React SPA
-    [Documentation]    Compare collectors on a React single-page application
-    [Tags]    collector    compare    spa
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://react-shopping-cart-67954.firebaseapp.com/
-    
-    # Collect with JSQuery
-    ${js_elements}=    JSQueryConnector.Collect UI Candidates
-    Log To Console    \n=== COMPARISON: React Shopping Cart SPA ===
-    Log To Console    JSQuery found: ${js_elements.__len__()} elements
-    
-    # Collect with AXTree
-    ${ax_elements}=    AXTreeConnector.Collect UI Candidates
-    Log To Console    AXTree found: ${ax_elements.__len__()} elements
-    
-    # Compare counts
-    ${diff}=    Evaluate    ${js_elements.__len__()} - ${ax_elements.__len__()}
-    Log To Console    Difference: ${diff} elements
-    
-    # Log element samples
-    Log To Console    \n=== JSQuery Sample (first 3) ===
-    FOR    ${element}    IN    @{js_elements[:3]}
-        Log To Console    - ${element['class_name']}: ${element['text'][:50]} | role: ${element['role']}
-    END
-    
-    Log To Console    \n=== AXTree Sample (first 3) ===
-    FOR    ${element}    IN    @{ax_elements[:3]}
-        Log To Console    - ${element['class_name']}: ${element['text'][:50]} | role: ${element['role']}
-    END
-    
-    Should Not Be Empty    ${js_elements}
-    Should Not Be Empty    ${ax_elements}
-    
-    [Teardown]    Close Browser
-
-
-*** Keywords ***
-Count Element Type
-    [Documentation]    Count how many elements of a specific tag type are in the list
-    [Arguments]    ${elements}    ${tag_name}
-    
-    ${count}=    Set Variable    0
-    FOR    ${element}    IN    @{elements}
-        ${class_name}=    Get From Dictionary    ${element}    class_name
-        ${count}=    Run Keyword If    '${class_name}' == '${tag_name}'
-        ...    Evaluate    ${count} + 1
-        ...    ELSE    Set Variable    ${count}
-    END
-    
-    RETURN    ${count}
-
diff --git a/tests/atest/compounent/generate_tests.py b/tests/atest/compounent/generate_tests.py
deleted file mode 100644
index 5400c94..0000000
--- a/tests/atest/compounent/generate_tests.py
+++ /dev/null
@@ -1,119 +0,0 @@
-#!/usr/bin/env python
-"""
-Generate Robot Framework tests from websites_100.yaml
-
-This script generates 100 individual test cases (no template, no FOR loop)
-from the YAML file containing website URLs.
-
-Usage:
-    python generate_tests.py > test_100_websites.robot
-    
-Or directly:
-    python generate_tests.py
-"""
-
-import yaml
-import sys
-
-
-def load_websites(yaml_file='websites_100.yaml'):
-    """Load websites from YAML file."""
-    with open(yaml_file, 'r') as f:
-        data = yaml.safe_load(f)
-    return data['websites']
-
-
-def generate_test_case(site):
-    """
-    Generate a single Robot Framework test case.
-    
-    Each test is exactly the same structure:
-    - Open Website
-    - Collect Elements
-    - Take Screenshot  
-    - Annotate with BBox
-    """
-    test = f"""
-Test {site['name']}
-    [Documentation]    Category: {site['category']} | URL: {site['url']}
-    [Tags]    {site['category']}    100websites
-    
-    Open Website    {site['url']}
-    
-    ${{elements}}=    Connector.Collect UI Candidates
-    Log Many    @{{elements}}
-    
-    ${{screenshot}}=    Take Screenshot    filename={site['name']}.png    fullPage=True
-    ${{annotated}}=    Draw And Log Elements    ${{elements}}    ${{screenshot}}
-    
-    [Teardown]    Close Browser
-"""
-    return test
-
-
-def generate_robot_file(websites):
-    """Generate complete Robot Framework file with 100 tests."""
-    
-    header = """*** Settings ***
-Documentation    Test JSQuery collector on 100 diverse websites
-...              This file is AUTO-GENERATED by generate_tests.py
-...              Categories: demo(20), simple(20), text(15), images(15), spa(15), social(10), ecommerce(5)
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    js_query    AS    Connector
-Library          tests/atest/compounent/BBoxVisualizer.py
-Resource         ${EXECDIR}/tests/atest/browsers/config.robot
-
-Suite Setup      Log    Starting test suite on 100 websites
-Suite Teardown   Log    ✅ Test suite complete
-
-*** Test Cases ***"""
-    
-    # Generate all test cases
-    tests = [generate_test_case(site) for site in websites]
-    
-    # Combine header and tests
-    content = header + ''.join(tests)
-    
-    return content
-
-
-def main():
-    """Main function."""
-    print("Loading websites from websites_100.yaml...")
-    websites = load_websites()
-    
-    print(f"Found {len(websites)} websites")
-    
-    # Count by category
-    categories = {}
-    for site in websites:
-        cat = site['category']
-        categories[cat] = categories.get(cat, 0) + 1
-    
-    print("Distribution:")
-    for cat, count in sorted(categories.items()):
-        print(f"  {cat}: {count}")
-    
-    print("\nGenerating Robot Framework file...")
-    content = generate_robot_file(websites)
-    
-    # Write to file
-    output_file = 'test_100_websites.robot'
-    with open(output_file, 'w') as f:
-        f.write(content)
-    
-    print(f"✅ Generated {output_file}")
-    print(f"   Total lines: {len(content.splitlines())}")
-    print(f"   Total tests: {len(websites)}")
-    
-    return 0
-
-
-if __name__ == '__main__':
-    sys.exit(main())
-
-
-
-
-
-
diff --git a/tests/atest/compounent/test_axtree.robot b/tests/atest/compounent/test_axtree.robot
deleted file mode 100644
index ee381c2..0000000
--- a/tests/atest/compounent/test_axtree.robot
+++ /dev/null
@@ -1,100 +0,0 @@
-*** Settings ***
-Documentation    Test AXTree collector strategy for UI element collection
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    axtree    AS    Connector
-Library          tests/atest/compounent/BBoxVisualizer.py
-
-
-*** Test Cases ***
-Test AXTree Collector On Simple Login Form
-    [Documentation]    Test AXTree collector on a simple login form
-    [Tags]    collector    axtree    login
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log    \n=== AXTREE COLLECTOR ===
-    Log    Found ${elements.__len__()} elements
-    
-    Log Many    @{elements}
-    
-    Should Not Be Empty    ${elements}
-    Should Be True    ${elements.__len__()} >= 3
-    
-    [Teardown]    Close Browser
-
-
-Test AXTree Collector On SauceDemo
-    [Documentation]    Test AXTree collector on SauceDemo login page
-    [Tags]    collector    axtree    saucedemo
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://www.saucedemo.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log    \n=== AXTREE COLLECTOR (SauceDemo) ===
-    Log    Found ${elements.__len__()} elements
-    
-    Log Many    @{elements}
-    
-    # Should find at least username, password, and login button
-    Should Not Be Empty    ${elements}
-    Should Be True    ${elements.__len__()} >= 3
-    
-    [Teardown]    Close Browser
-
-
-Test AXTree Collector Element Details
-    [Documentation]    Verify that AXTree collector captures all required attributes
-    [Tags]    collector    axtree    details
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    ${elements}=    Connector.Collect UI Candidates
-    
-    # Verify first element has expected keys
-    ${first_element}=    Set Variable    ${elements}[0]
-    Should Contain    ${first_element}    text
-    Should Contain    ${first_element}    resource_id
-    Should Contain    ${first_element}    content_desc
-    Should Contain    ${first_element}    class_name
-    Should Contain    ${first_element}    role
-    Should Contain    ${first_element}    clickable
-    Should Contain    ${first_element}    enabled
-    
-    Log    \n=== First Element Details ===
-    Log    ${first_element}
-    
-    [Teardown]    Close Browser
-
-
-Test AXTree With BBox Visualization
-    [Documentation]    Visualize AXTree bounding boxes on screenshot
-    [Tags]    collector    axtree    visualization    bbox
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    # Collect elements
-    ${elements}=    Connector.Collect UI Candidates
-    Log    Found ${elements.__len__()} accessible elements with bounding boxes
-    
-    # Take screenshot
-    ${screenshot}=    Take Screenshot    filename=axtree_original.png    fullPage=True
-    
-    # Draw bounding boxes and embed in log
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    Log    ✅ AXTree annotated screenshot saved: ${annotated}
-    Log    📦 Total accessible elements annotated: ${elements.__len__()}
-    
-    Should Not Be Empty    ${annotated}
-    
-    [Teardown]    Close Browser
-
diff --git a/tests/atest/compounent/test_bbox.py b/tests/atest/compounent/test_bbox.py
deleted file mode 100644
index 1e55b73..0000000
--- a/tests/atest/compounent/test_bbox.py
+++ /dev/null
@@ -1,172 +0,0 @@
-#!/usr/bin/env python
-"""
-Test script to verify that bounding boxes are correctly returned by collectors.
-
-This script tests that:
-1. bbox field exists in collected elements
-2. bbox contains x, y, width, height
-3. bbox values are valid integers
-4. bbox values make sense (positive, reasonable ranges)
-"""
-
-import unittest
-from Agent.platforms.collectors.js_query_collector import JSQueryCollector
-from Agent.platforms.collectors.axtree_collector import AXTreeCollector
-
-
-class TestBBoxStructure(unittest.TestCase):
-    """Test that bbox structure is correct (no browser needed)."""
-    
-    def test_jsquery_collector_has_bbox_in_mapping(self):
-        """Test that JSQueryCollector maps bbox field."""
-        collector = JSQueryCollector()
-        
-        # Simulate data returned from JavaScript
-        mock_elem_data = {
-            'text': 'Click me',
-            'id': 'btn-1',
-            'tag': 'button',
-            'bbox': {'x': 100, 'y': 200, 'width': 80, 'height': 32}
-        }
-        
-        # This is what the collector does internally
-        attrs = {
-            'text': mock_elem_data.get('text', ''),
-            'resource_id': mock_elem_data.get('testId') or mock_elem_data.get('id', ''),
-            'content_desc': mock_elem_data.get('ariaLabel') or mock_elem_data.get('placeholder', ''),
-            'label': mock_elem_data.get('label', ''),
-            'class_name': mock_elem_data.get('tag', ''),
-            'role': mock_elem_data.get('role', ''),
-            'name': mock_elem_data.get('name', ''),
-            'type': mock_elem_data.get('type', ''),
-            'href': mock_elem_data.get('href', ''),
-            'clickable': True,
-            'enabled': True,
-            'bbox': mock_elem_data.get('bbox', {}),
-        }
-        
-        # Verify bbox is present and correct
-        self.assertIn('bbox', attrs)
-        self.assertEqual(attrs['bbox'], {'x': 100, 'y': 200, 'width': 80, 'height': 32})
-    
-    def test_axtree_collector_has_bbox_in_mapping(self):
-        """Test that AXTreeCollector maps bbox field."""
-        collector = AXTreeCollector()
-        
-        # Simulate data returned from JavaScript
-        mock_elem_data = {
-            'text': 'Submit',
-            'id': 'submit-btn',
-            'tag': 'button',
-            'role': 'button',
-            'accessibleName': 'Submit form',
-            'bbox': {'x': 50, 'y': 400, 'width': 100, 'height': 40}
-        }
-        
-        # This is what the collector does internally
-        accessible_name = mock_elem_data.get('accessibleName', '')
-        text = mock_elem_data.get('text', '')
-        
-        attrs = {
-            'text': text,
-            'resource_id': mock_elem_data.get('testId') or mock_elem_data.get('id', ''),
-            'content_desc': accessible_name or mock_elem_data.get('placeholder', ''),
-            'label': accessible_name if accessible_name != text else '',
-            'class_name': mock_elem_data.get('tag', ''),
-            'role': mock_elem_data.get('role', ''),
-            'name': mock_elem_data.get('name', ''),
-            'type': mock_elem_data.get('type', ''),
-            'href': mock_elem_data.get('href', ''),
-            'clickable': True,
-            'enabled': True,
-            'bbox': mock_elem_data.get('bbox', {}),
-        }
-        
-        # Verify bbox is present and correct
-        self.assertIn('bbox', attrs)
-        self.assertEqual(attrs['bbox'], {'x': 50, 'y': 400, 'width': 100, 'height': 40})
-    
-    def test_bbox_structure(self):
-        """Test that bbox has the correct structure."""
-        bbox = {'x': 100, 'y': 200, 'width': 80, 'height': 32}
-        
-        # Verify all required keys are present
-        self.assertIn('x', bbox)
-        self.assertIn('y', bbox)
-        self.assertIn('width', bbox)
-        self.assertIn('height', bbox)
-        
-        # Verify all values are integers
-        self.assertIsInstance(bbox['x'], int)
-        self.assertIsInstance(bbox['y'], int)
-        self.assertIsInstance(bbox['width'], int)
-        self.assertIsInstance(bbox['height'], int)
-    
-    def test_bbox_values_are_positive(self):
-        """Test that bbox values are positive (or zero)."""
-        bbox = {'x': 100, 'y': 200, 'width': 80, 'height': 32}
-        
-        self.assertGreaterEqual(bbox['x'], 0)
-        self.assertGreaterEqual(bbox['y'], 0)
-        self.assertGreater(bbox['width'], 0)  # width must be > 0 for visible elements
-        self.assertGreater(bbox['height'], 0)  # height must be > 0 for visible elements
-
-
-class TestBBoxDocumentation(unittest.TestCase):
-    """Test that bbox is documented in BaseUICollector."""
-    
-    def test_base_collector_documents_bbox(self):
-        """Test that BaseUICollector docstring mentions bbox."""
-        from Agent.platforms.collectors.base_collector import BaseUICollector
-        
-        docstring = BaseUICollector.collect_elements.__doc__
-        self.assertIsNotNone(docstring)
-        self.assertIn('bbox', docstring.lower())
-
-
-def validate_bbox(bbox: dict) -> tuple[bool, str]:
-    """
-    Validate a bbox dictionary.
-    
-    Returns:
-        (is_valid, error_message)
-    """
-    if not isinstance(bbox, dict):
-        return False, f"bbox is not a dict, got {type(bbox)}"
-    
-    required_keys = ['x', 'y', 'width', 'height']
-    for key in required_keys:
-        if key not in bbox:
-            return False, f"bbox missing key: {key}"
-    
-    for key in required_keys:
-        if not isinstance(bbox[key], int):
-            return False, f"bbox[{key}] is not an int, got {type(bbox[key])}"
-    
-    if bbox['width'] <= 0:
-        return False, f"bbox width must be > 0, got {bbox['width']}"
-    
-    if bbox['height'] <= 0:
-        return False, f"bbox height must be > 0, got {bbox['height']}"
-    
-    # x and y can be negative (element off-screen to the left/top)
-    # but let's warn if they're too negative
-    if bbox['x'] < -10000 or bbox['y'] < -10000:
-        return False, f"bbox coordinates seem unrealistic: x={bbox['x']}, y={bbox['y']}"
-    
-    # Reasonable max dimensions (8K screen is ~7680 x 4320)
-    if bbox['width'] > 20000 or bbox['height'] > 20000:
-        return False, f"bbox dimensions seem too large: {bbox['width']}x{bbox['height']}"
-    
-    return True, "OK"
-
-
-if __name__ == "__main__":
-    print("=" * 60)
-    print("Testing BBox Implementation")
-    print("=" * 60)
-    print()
-    
-    # Run unit tests
-    unittest.main(verbosity=2)
-
diff --git a/tests/atest/compounent/test_bbox_visualizer.py b/tests/atest/compounent/test_bbox_visualizer.py
deleted file mode 100644
index a2968d8..0000000
--- a/tests/atest/compounent/test_bbox_visualizer.py
+++ /dev/null
@@ -1,261 +0,0 @@
-#!/usr/bin/env python
-"""
-Unit tests for BBoxVisualizer library.
-
-These tests verify the BBoxVisualizer functionality without requiring
-a browser or real screenshots.
-"""
-
-import unittest
-import os
-import tempfile
-from PIL import Image
-from tests.atest.compounent.BBoxVisualizer import BBoxVisualizer
-
-
-class TestBBoxVisualizerUnit(unittest.TestCase):
-    """Unit tests for BBoxVisualizer."""
-    
-    def setUp(self):
-        """Set up test fixtures."""
-        self.visualizer = BBoxVisualizer()
-        self.temp_dir = tempfile.mkdtemp()
-        
-        # Create a simple test image (white background)
-        self.test_image_path = os.path.join(self.temp_dir, "test.png")
-        img = Image.new('RGB', (800, 600), 'white')
-        img.save(self.test_image_path)
-        
-        # Sample elements with bbox
-        self.sample_elements = [
-            {
-                'text': 'Username',
-                'resource_id': 'username',
-                'class_name': 'input',
-                'bbox': {'x': 155, 'y': 202, 'width': 470, 'height': 32}
-            },
-            {
-                'text': 'Password',
-                'resource_id': 'password',
-                'class_name': 'input',
-                'bbox': {'x': 155, 'y': 265, 'width': 470, 'height': 32}
-            },
-            {
-                'text': 'Login',
-                'class_name': 'button',
-                'bbox': {'x': 155, 'y': 312, 'width': 162, 'height': 59}
-            }
-        ]
-    
-    def test_library_initialization(self):
-        """Test that BBoxVisualizer can be initialized."""
-        visualizer = BBoxVisualizer()
-        self.assertIsNotNone(visualizer)
-        self.assertEqual(visualizer.default_color, 'red')
-        self.assertEqual(visualizer.default_width, 3)
-    
-    def test_draw_bbox_on_screenshot_returns_path(self):
-        """Test that draw_bbox_on_screenshot returns a valid path."""
-        output_path = self.visualizer.draw_bbox_on_screenshot(
-            self.test_image_path,
-            self.sample_elements
-        )
-        
-        self.assertIsNotNone(output_path)
-        self.assertTrue(os.path.exists(output_path))
-        self.assertTrue(output_path.endswith('_annotated.png'))
-    
-    def test_draw_bbox_creates_valid_image(self):
-        """Test that the annotated image is valid."""
-        output_path = self.visualizer.draw_bbox_on_screenshot(
-            self.test_image_path,
-            self.sample_elements
-        )
-        
-        # Verify we can open the annotated image
-        img = Image.open(output_path)
-        self.assertEqual(img.size, (800, 600))
-        self.assertEqual(img.mode, 'RGB')
-    
-    def test_draw_bbox_with_custom_output_path(self):
-        """Test draw_bbox with custom output path."""
-        custom_output = os.path.join(self.temp_dir, 'custom_annotated.png')
-        
-        output_path = self.visualizer.draw_bbox_on_screenshot(
-            self.test_image_path,
-            self.sample_elements,
-            output_path=custom_output
-        )
-        
-        self.assertEqual(output_path, custom_output)
-        self.assertTrue(os.path.exists(custom_output))
-    
-    def test_draw_bbox_with_empty_elements(self):
-        """Test draw_bbox with no elements."""
-        output_path = self.visualizer.draw_bbox_on_screenshot(
-            self.test_image_path,
-            []
-        )
-        
-        # Should still create an image (just no annotations)
-        self.assertTrue(os.path.exists(output_path))
-    
-    def test_draw_bbox_with_missing_bbox(self):
-        """Test draw_bbox with elements missing bbox."""
-        elements_no_bbox = [
-            {'text': 'Element without bbox', 'class_name': 'button'}
-        ]
-        
-        output_path = self.visualizer.draw_bbox_on_screenshot(
-            self.test_image_path,
-            elements_no_bbox
-        )
-        
-        # Should handle gracefully
-        self.assertTrue(os.path.exists(output_path))
-    
-    def test_draw_bbox_with_custom_color(self):
-        """Test draw_bbox with custom color."""
-        output_path = self.visualizer.draw_bbox_on_screenshot(
-            self.test_image_path,
-            self.sample_elements,
-            color='blue'
-        )
-        
-        self.assertTrue(os.path.exists(output_path))
-    
-    def test_annotate_elements_creates_timestamped_file(self):
-        """Test that annotate_elements creates a timestamped file."""
-        output_dir = os.path.join(self.temp_dir, 'annotated')
-        
-        output_path = self.visualizer.annotate_elements(
-            self.sample_elements,
-            self.test_image_path,
-            output_dir=output_dir
-        )
-        
-        # Verify output directory was created
-        self.assertTrue(os.path.exists(output_dir))
-        
-        # Verify file has timestamp
-        self.assertTrue('_annotated_' in output_path)
-        self.assertTrue(os.path.exists(output_path))
-    
-    def test_library_metadata(self):
-        """Test that library metadata is correct."""
-        self.assertEqual(BBoxVisualizer.ROBOT_LIBRARY_SCOPE, 'GLOBAL')
-        self.assertIsNotNone(BBoxVisualizer.ROBOT_LIBRARY_VERSION)
-    
-    def tearDown(self):
-        """Clean up test fixtures."""
-        # Clean up temporary files
-        import shutil
-        if os.path.exists(self.temp_dir):
-            shutil.rmtree(self.temp_dir)
-
-
-class TestBBoxVisualizerIntegration(unittest.TestCase):
-    """Integration tests for BBoxVisualizer with real bbox data."""
-    
-    def test_visualize_login_form(self):
-        """Test visualizing a login form (from actual collector output)."""
-        visualizer = BBoxVisualizer()
-        
-        # Real data from collector output (from user's example)
-        elements = [
-            {
-                'text': '',
-                'resource_id': 'username',
-                'content_desc': '',
-                'label': 'Username',
-                'class_name': 'input',
-                'role': '',
-                'name': 'username',
-                'type': 'text',
-                'href': '',
-                'clickable': True,
-                'enabled': True,
-                'bbox': {'x': 155, 'y': 202, 'width': 470, 'height': 32}
-            },
-            {
-                'text': '',
-                'resource_id': 'password',
-                'content_desc': '',
-                'label': 'Password',
-                'class_name': 'input',
-                'role': '',
-                'name': 'password',
-                'type': 'password',
-                'href': '',
-                'clickable': True,
-                'enabled': True,
-                'bbox': {'x': 155, 'y': 265, 'width': 470, 'height': 32}
-            },
-            {
-                'text': 'Login',
-                'resource_id': '',
-                'content_desc': '',
-                'label': '',
-                'class_name': 'button',
-                'role': '',
-                'name': '',
-                'type': 'submit',
-                'href': '',
-                'clickable': True,
-                'enabled': True,
-                'bbox': {'x': 155, 'y': 312, 'width': 162, 'height': 59}
-            },
-            {
-                'text': 'Elemental Selenium',
-                'resource_id': '',
-                'content_desc': '',
-                'label': '',
-                'class_name': 'a',
-                'role': '',
-                'name': '',
-                'type': '',
-                'href': 'http://elementalselenium.com/',
-                'clickable': True,
-                'enabled': True,
-                'bbox': {'x': 614, 'y': 446, 'width': 141, 'height': 18}
-            }
-        ]
-        
-        # Create a mock screenshot
-        temp_dir = tempfile.mkdtemp()
-        screenshot_path = os.path.join(temp_dir, 'login_page.png')
-        img = Image.new('RGB', (1024, 768), 'white')
-        img.save(screenshot_path)
-        
-        # Annotate
-        output_path = visualizer.draw_bbox_on_screenshot(
-            screenshot_path,
-            elements
-        )
-        
-        # Verify
-        self.assertTrue(os.path.exists(output_path))
-        
-        # Verify all 4 elements were annotated
-        annotated_img = Image.open(output_path)
-        self.assertIsNotNone(annotated_img)
-        
-        # Cleanup
-        import shutil
-        shutil.rmtree(temp_dir)
-
-
-if __name__ == "__main__":
-    print("=" * 60)
-    print("Testing BBoxVisualizer Library")
-    print("=" * 60)
-    print()
-    
-    # Run tests
-    unittest.main(verbosity=2)
-
-
-
-
-
-
diff --git a/tests/atest/compounent/test_compare_visualization.robot b/tests/atest/compounent/test_compare_visualization.robot
deleted file mode 100644
index 8ff273d..0000000
--- a/tests/atest/compounent/test_compare_visualization.robot
+++ /dev/null
@@ -1,85 +0,0 @@
-*** Settings ***
-Documentation    Compare JSQuery and AXTree collectors with side-by-side visualization
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    js_query    AS    JSConnector
-Library          Agent.platforms._webconnector.WebConnectorRF    axtree    AS    AXConnector
-Library          tests/atest/compounent/BBoxVisualizer.py
-
-
-*** Test Cases ***
-Compare JSQuery vs AXTree With Visualization
-    [Documentation]    Visual comparison of two collectors on the same page
-    [Tags]    collector    comparison    visualization    bbox
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    # Collect with both collectors
-    ${js_elements}=    JSConnector.Collect UI Candidates
-    ${ax_elements}=    AXConnector.Collect UI Candidates
-    
-    Log    JSQuery found: ${js_elements.__len__()} elements
-    Log    AXTree found: ${ax_elements.__len__()} elements
-    
-    # Take screenshot
-    ${screenshot}=    Take Screenshot    filename=comparison.png    fullPage=True
-    
-    # Create comparison image
-    ${comparison}=    Create Comparison Image
-    ...    ${screenshot}
-    ...    ${js_elements}
-    ...    ${ax_elements}
-    ...    JSQuery (${js_elements.__len__()})
-    ...    AXTree (${ax_elements.__len__()})
-    
-    Log    📊 Comparison image created: ${comparison}
-    
-    # Also create individual annotated versions
-    ${js_annotated}=    Draw BBox On Screenshot
-    ...    ${screenshot}
-    ...    ${js_elements}
-    ...    color=red
-    
-    ${ax_annotated}=    Draw BBox On Screenshot
-    ...    ${screenshot}
-    ...    ${ax_elements}
-    ...    color=blue
-    
-    Log    🔴 JSQuery only: ${js_annotated}
-    Log    🔵 AXTree only: ${ax_annotated}
-    
-    [Teardown]    Close Browser
-
-
-Compare On Complex Page
-    [Documentation]    Compare collectors on a more complex page (SauceDemo)
-    [Tags]    collector    comparison    visualization    bbox    complex
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://www.saucedemo.com
-    
-    # Collect with both collectors
-    ${js_elements}=    JSConnector.Collect UI Candidates
-    ${ax_elements}=    AXConnector.Collect UI Candidates
-    
-    Log    📊 JSQuery: ${js_elements.__len__()} elements
-    Log    📊 AXTree: ${ax_elements.__len__()} elements
-    
-    ${diff}=    Evaluate    ${js_elements.__len__()} - ${ax_elements.__len__()}
-    Log    📊 Difference: ${diff} elements
-    
-    # Take screenshot and create comparison
-    ${screenshot}=    Take Screenshot    filename=saucedemo_compare.png    fullPage=True
-    ${comparison}=    Create Comparison Image
-    ...    ${screenshot}
-    ...    ${js_elements}
-    ...    ${ax_elements}
-    ...    JSQuery
-    ...    AXTree
-    
-    Log    Comparison saved: ${comparison}
-    
-    [Teardown]    Close Browser
-
diff --git a/tests/atest/compounent/test_js_query.robot b/tests/atest/compounent/test_js_query.robot
deleted file mode 100644
index 1c5050e..0000000
--- a/tests/atest/compounent/test_js_query.robot
+++ /dev/null
@@ -1,115 +0,0 @@
-*** Settings ***
-Documentation    Test JSQuery collector strategy for UI element collection
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    js_query    AS    Connector
-Library          tests/atest/compounent/BBoxVisualizer.py
-Resource         ${EXECDIR}/tests/atest/browsers/config.robot
-
-*** Test Cases ***
-Test Herokuapp
-    [Documentation]    Test JSQuery collector on a simple login form
-    [Tags]    collector    js_query    login
-    
-    Open Website    https://the-internet.herokuapp.com/login
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-
-    ${screenshot}=    Take Screenshot    filename=original.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-
-Test JSQuery Collector On SauceDemo
-    [Documentation]    Test JSQuery collector on SauceDemo login page
-    [Tags]    collector    js_query    saucedemo
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://www.saucedemo.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log    \n=== JS_QUERY COLLECTOR (SauceDemo) ===
-    Log    Found ${elements.__len__()} elements
-    
-    Log Many    @{elements}
-    
-    # Should find at least username, password, and login button
-    Should Not Be Empty    ${elements}
-    Should Be True    ${elements.__len__()} >= 3
-    
-    [Teardown]    Close Browser
-
-
-Test JSQuery Collector Element Details
-    [Documentation]    Verify that JSQuery collector captures all required attributes
-    [Tags]    collector    js_query    details
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    ${elements}=    Connector.Collect UI Candidates
-    
-    # Verify first element has expected keys
-    ${first_element}=    Set Variable    ${elements}[0]
-    Should Contain    ${first_element}    text
-    Should Contain    ${first_element}    resource_id
-    Should Contain    ${first_element}    content_desc
-    Should Contain    ${first_element}    class_name
-    Should Contain    ${first_element}    clickable
-    Should Contain    ${first_element}    enabled
-    
-    Log    \n=== First Element Details ===
-    Log    ${first_element}
-    
-    [Teardown]    Close Browser
-
-
-Test JSQuery With BBox Visualization
-    [Documentation]    Visualize bounding boxes on screenshot with annotations
-    [Tags]    collector    js_query    visualization    bbox
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://the-internet.herokuapp.com/login
-    
-    # Collect elements
-    ${elements}=    Connector.Collect UI Candidates
-    Log    Found ${elements.__len__()} elements with bounding boxes
-    
-    # Take screenshot
-    ${screenshot}=    Take Screenshot    filename=original.png    fullPage=True
-    
-    # Draw bounding boxes and embed in log
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    Log    ✅ Annotated screenshot saved: ${annotated}
-    Log    📦 Total elements annotated: ${elements.__len__()}
-    
-    Should Not Be Empty    ${annotated}
-    
-    [Teardown]    Close Browser
-
-
-Test JSQuery BBox On SauceDemo
-    [Documentation]    Test BBox visualization on SauceDemo login page
-    [Tags]    collector    js_query    visualization    bbox    saucedemo
-    
-    New Browser    chromium    headless=False
-    New Context
-    New Page    https://www.saucedemo.com
-    
-    # Collect elements
-    ${elements}=    Connector.Collect UI Candidates
-    Log    Found ${elements.__len__()} elements
-    
-    # Take screenshot and annotate
-    ${screenshot}=    Take Screenshot    filename=saucedemo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}    width=1000
-    
-    Log    Annotated: ${annotated}
-    
-    [Teardown]    Close Browser
-
diff --git a/tests/atest/compounent/test_quick_check.robot b/tests/atest/compounent/test_quick_check.robot
deleted file mode 100644
index d5c124e..0000000
--- a/tests/atest/compounent/test_quick_check.robot
+++ /dev/null
@@ -1,38 +0,0 @@
-*** Settings ***
-Documentation    Quick smoke test to verify collector system works
-Library          Agent.platforms._webconnector.WebConnectorRF    js_query    AS    JSQueryConn
-Library          Agent.platforms._webconnector.WebConnectorRF    axtree    AS    AXTreeConn
-
-
-*** Test Cases ***
-Verify JSQuery Connector Works
-    [Documentation]    Test that JSQuery connector can be instantiated
-    [Tags]    smoke    js_query
-    
-    ${strategy}=    Get Variable Value    ${JSQueryConn.collector_strategy}
-    Should Be Equal    ${strategy}    js_query
-    Log To Console    \n✓ JSQuery connector created with strategy: ${strategy}
-
-
-Verify AXTree Connector Works
-    [Documentation]    Test that AXTree connector can be instantiated
-    [Tags]    smoke    axtree
-    
-    ${strategy}=    Get Variable Value    ${AXTreeConn.collector_strategy}
-    Should Be Equal    ${strategy}    axtree
-    Log To Console    ✓ AXTree connector created with strategy: ${strategy}
-
-
-Verify Different Collectors Are Independent
-    [Documentation]    Test that two connectors with different strategies are separate instances
-    [Tags]    smoke
-    
-    ${js_strategy}=    Get Variable Value    ${JSQueryConn.collector_strategy}
-    ${ax_strategy}=    Get Variable Value    ${AXTreeConn.collector_strategy}
-    
-    Should Be Equal    ${js_strategy}    js_query
-    Should Be Equal    ${ax_strategy}    axtree
-    Should Not Be Equal    ${js_strategy}    ${ax_strategy}
-    
-    Log To Console    \n✓ Multiple independent collectors work correctly
-
diff --git a/tests/atest/compounent/test_web_ui_acctree.robot b/tests/atest/compounent/test_web_ui_acctree.robot
deleted file mode 100644
index 96d3292..0000000
--- a/tests/atest/compounent/test_web_ui_acctree.robot
+++ /dev/null
@@ -1,1398 +0,0 @@
-*** Settings ***
-Documentation    Test JSQuery collector on 100 diverse websites
-...              This file is AUTO-GENERATED by generate_tests.py
-...              Categories: demo(20), simple(20), text(15), images(15), spa(15), social(10), ecommerce(5)
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    axtree    AS    Connector
-Library          tests/atest/compounent/BBoxVisualizer.py
-Resource         ${EXECDIR}/tests/atest/browsers/config.robot
-
-Suite Setup      Log    Starting test suite on 100 websites
-Suite Teardown   Log    ✅ Test suite complete
-
-*** Test Cases ***
-Test HerokuApp_Login
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/login
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/login
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Login.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test SauceDemo
-    [Documentation]    Category: demo | URL: https://www.saucedemo.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://www.saucedemo.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=SauceDemo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test DemoQA
-    [Documentation]    Category: demo | URL: https://demoqa.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://demoqa.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=DemoQA.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Dropdown
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/dropdown
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/dropdown
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Dropdown.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Checkboxes
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/checkboxes
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/checkboxes
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Checkboxes.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Inputs
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/inputs
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/inputs
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Inputs.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Upload
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/upload
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/upload
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Upload.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Tables
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/tables
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/tables
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Tables.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_IFrame
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/iframe
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/iframe
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_IFrame.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_DynamicLoading
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/dynamic_loading/1
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/dynamic_loading/1
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_DynamicLoading.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-# Test AutomationExercise
-#     [Documentation]    Category: demo | URL: https://automationexercise.com
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://automationexercise.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=AutomationExercise.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test XYZ_Bank
-#     [Documentation]    Category: demo | URL: https://www.globalsqa.com/angularJs-protractor/BankingProject
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://www.globalsqa.com/angularJs-protractor/BankingProject
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=XYZ_Bank.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test PHPTravels_Demo
-#     [Documentation]    Category: demo | URL: https://phptravels.com/demo
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://phptravels.com/demo
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=PHPTravels_Demo.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test OpenCart_Demo
-#     [Documentation]    Category: demo | URL: https://demo.opencart.com
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://demo.opencart.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=OpenCart_Demo.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test NopCommerce_Demo
-#     [Documentation]    Category: demo | URL: https://demo.nopcommerce.com
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://demo.nopcommerce.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=NopCommerce_Demo.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Magento_Demo
-#     [Documentation]    Category: demo | URL: https://magento.softwaretestingboard.com
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://magento.softwaretestingboard.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Magento_Demo.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test ExpandTesting_Practice
-#     [Documentation]    Category: demo | URL: https://practice.expandtesting.com
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://practice.expandtesting.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=ExpandTesting_Practice.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test EvilTester_Pages
-#     [Documentation]    Category: demo | URL: https://testpages.eviltester.com/styled/index.html
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://testpages.eviltester.com/styled/index.html
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=EvilTester_Pages.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test AutomationCamp
-#     [Documentation]    Category: demo | URL: https://play1.automationcamp.ir/index.html
-#     [Tags]    demo    100websites
-    
-#     Open Website    https://play1.automationcamp.ir/index.html
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=AutomationCamp.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Example
-#     [Documentation]    Category: simple | URL: https://example.com
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://example.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Example.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test CERN_FirstWebsite
-#     [Documentation]    Category: simple | URL: https://info.cern.ch
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://info.cern.ch
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=CERN_FirstWebsite.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test W3C
-#     [Documentation]    Category: simple | URL: https://www.w3.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.w3.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=W3C.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test IANA
-#     [Documentation]    Category: simple | URL: https://www.iana.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.iana.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=IANA.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test RFC_Editor
-#     [Documentation]    Category: simple | URL: https://www.rfc-editor.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.rfc-editor.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=RFC_Editor.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Kernel_Org
-#     [Documentation]    Category: simple | URL: https://www.kernel.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.kernel.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Kernel_Org.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Debian
-#     [Documentation]    Category: simple | URL: https://www.debian.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.debian.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Debian.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Python_Org
-#     [Documentation]    Category: simple | URL: https://www.python.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.python.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Python_Org.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test NPM
-#     [Documentation]    Category: simple | URL: https://www.npmjs.com
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.npmjs.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=NPM.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test PyPI
-#     [Documentation]    Category: simple | URL: https://pypi.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://pypi.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=PyPI.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Rust_Lang
-#     [Documentation]    Category: simple | URL: https://www.rust-lang.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.rust-lang.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Rust_Lang.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Go_Lang
-#     [Documentation]    Category: simple | URL: https://golang.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://golang.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Go_Lang.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Ruby_Lang
-#     [Documentation]    Category: simple | URL: https://www.ruby-lang.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.ruby-lang.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Ruby_Lang.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test PHP_Net
-#     [Documentation]    Category: simple | URL: https://www.php.net
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.php.net
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=PHP_Net.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Perl_Org
-#     [Documentation]    Category: simple | URL: https://www.perl.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.perl.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Perl_Org.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Scala_Lang
-#     [Documentation]    Category: simple | URL: https://www.scala-lang.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.scala-lang.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Scala_Lang.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Kotlin_Lang
-#     [Documentation]    Category: simple | URL: https://kotlinlang.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://kotlinlang.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Kotlin_Lang.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test SQLite
-#     [Documentation]    Category: simple | URL: https://www.sqlite.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.sqlite.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=SQLite.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test PostgreSQL
-#     [Documentation]    Category: simple | URL: https://www.postgresql.org
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.postgresql.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=PostgreSQL.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Nginx
-#     [Documentation]    Category: simple | URL: https://www.nginx.com
-#     [Tags]    simple    100websites
-    
-#     Open Website    https://www.nginx.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Nginx.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_Python
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Python_(programming_language)
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Python_(programming_language)
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_Python.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_AI
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Artificial_intelligence
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Artificial_intelligence
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_AI.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_JavaScript
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/JavaScript
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/JavaScript
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_JavaScript.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_WebDev
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Web_development
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Web_development
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_WebDev.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_Testing
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Software_testing
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Software_testing
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_Testing.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_ML
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Machine_learning
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Machine_learning
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_ML.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_DataScience
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Data_science
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Data_science
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_DataScience.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_Cloud
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Cloud_computing
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Cloud_computing
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_Cloud.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_Blockchain
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Blockchain
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Blockchain
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_Blockchain.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Wikipedia_Cybersecurity
-#     [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Cybersecurity
-#     [Tags]    text    100websites
-    
-#     Open Website    https://en.wikipedia.org/wiki/Cybersecurity
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Wikipedia_Cybersecurity.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test BBC_News
-#     [Documentation]    Category: text | URL: https://www.bbc.com/news
-#     [Tags]    text    100websites
-    
-#     Open Website    https://www.bbc.com/news
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=BBC_News.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Guardian_Tech
-#     [Documentation]    Category: text | URL: https://www.theguardian.com/technology
-#     [Tags]    text    100websites
-    
-#     Open Website    https://www.theguardian.com/technology
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Guardian_Tech.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test DevTo
-#     [Documentation]    Category: text | URL: https://dev.to
-#     [Tags]    text    100websites
-    
-#     Open Website    https://dev.to
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=DevTo.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test FreeCodeCamp_News
-#     [Documentation]    Category: text | URL: https://www.freecodecamp.org/news
-#     [Tags]    text    100websites
-    
-#     Open Website    https://www.freecodecamp.org/news
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=FreeCodeCamp_News.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Smashing_Magazine
-#     [Documentation]    Category: text | URL: https://www.smashingmagazine.com
-#     [Tags]    text    100websites
-    
-#     Open Website    https://www.smashingmagazine.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Smashing_Magazine.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Unsplash
-#     [Documentation]    Category: images | URL: https://unsplash.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://unsplash.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Unsplash.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Pexels
-#     [Documentation]    Category: images | URL: https://www.pexels.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.pexels.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Pexels.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Pixabay
-#     [Documentation]    Category: images | URL: https://pixabay.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://pixabay.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Pixabay.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Flickr_Explore
-#     [Documentation]    Category: images | URL: https://www.flickr.com/explore
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.flickr.com/explore
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Flickr_Explore.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test 500px
-#     [Documentation]    Category: images | URL: https://500px.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://500px.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=500px.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Behance
-#     [Documentation]    Category: images | URL: https://www.behance.net
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.behance.net
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Behance.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Dribbble
-#     [Documentation]    Category: images | URL: https://dribbble.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://dribbble.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Dribbble.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Awwwards
-#     [Documentation]    Category: images | URL: https://www.awwwards.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.awwwards.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Awwwards.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test DeviantArt
-#     [Documentation]    Category: images | URL: https://www.deviantart.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.deviantart.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=DeviantArt.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test ArtStation
-#     [Documentation]    Category: images | URL: https://www.artstation.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.artstation.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=ArtStation.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Shutterstock
-#     [Documentation]    Category: images | URL: https://www.shutterstock.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.shutterstock.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Shutterstock.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Adobe_Stock
-#     [Documentation]    Category: images | URL: https://stock.adobe.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://stock.adobe.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Adobe_Stock.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Getty_Images
-#     [Documentation]    Category: images | URL: https://www.gettyimages.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.gettyimages.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Getty_Images.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test iStock
-#     [Documentation]    Category: images | URL: https://www.istockphoto.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://www.istockphoto.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=iStock.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Burst_Shopify
-#     [Documentation]    Category: images | URL: https://burst.shopify.com
-#     [Tags]    images    100websites
-    
-#     Open Website    https://burst.shopify.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Burst_Shopify.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test React_ShoppingCart
-#     [Documentation]    Category: spa | URL: https://react-shopping-cart-67954.firebaseapp.com
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://react-shopping-cart-67954.firebaseapp.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=React_ShoppingCart.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test TodoMVC_React
-#     [Documentation]    Category: spa | URL: https://todomvc.com/examples/react
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://todomvc.com/examples/react
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=TodoMVC_React.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test TodoMVC_Vue
-#     [Documentation]    Category: spa | URL: https://todomvc.com/examples/vue
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://todomvc.com/examples/vue
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=TodoMVC_Vue.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test TodoMVC_Angular
-#     [Documentation]    Category: spa | URL: https://todomvc.com/examples/angular
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://todomvc.com/examples/angular
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=TodoMVC_Angular.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test React_Homepage
-#     [Documentation]    Category: spa | URL: https://reactjs.org
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://reactjs.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=React_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Vue_Homepage
-#     [Documentation]    Category: spa | URL: https://vuejs.org
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://vuejs.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Vue_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Angular_Homepage
-#     [Documentation]    Category: spa | URL: https://angular.io
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://angular.io
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Angular_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Svelte_Homepage
-#     [Documentation]    Category: spa | URL: https://svelte.dev
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://svelte.dev
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Svelte_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test NextJS_Homepage
-#     [Documentation]    Category: spa | URL: https://nextjs.org
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://nextjs.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=NextJS_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test NuxtJS_Homepage
-#     [Documentation]    Category: spa | URL: https://nuxtjs.org
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://nuxtjs.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=NuxtJS_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Gatsby_Homepage
-#     [Documentation]    Category: spa | URL: https://www.gatsbyjs.com
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://www.gatsbyjs.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Gatsby_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Remix_Homepage
-#     [Documentation]    Category: spa | URL: https://remix.run
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://remix.run
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Remix_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Astro_Homepage
-#     [Documentation]    Category: spa | URL: https://astro.build
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://astro.build
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Astro_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Qwik_Homepage
-#     [Documentation]    Category: spa | URL: https://qwik.builder.io
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://qwik.builder.io
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Qwik_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Solid_Homepage
-#     [Documentation]    Category: spa | URL: https://solidjs.com
-#     [Tags]    spa    100websites
-    
-#     Open Website    https://solidjs.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Solid_Homepage.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test HackerNews
-#     [Documentation]    Category: social | URL: https://news.ycombinator.com
-#     [Tags]    social    100websites
-    
-#     Open Website    https://news.ycombinator.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=HackerNews.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Reddit_Programming
-#     [Documentation]    Category: social | URL: https://www.reddit.com/r/programming
-#     [Tags]    social    100websites
-    
-#     Open Website    https://www.reddit.com/r/programming
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Reddit_Programming.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Reddit_WebDev
-#     [Documentation]    Category: social | URL: https://www.reddit.com/r/webdev
-#     [Tags]    social    100websites
-    
-#     Open Website    https://www.reddit.com/r/webdev
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Reddit_WebDev.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test StackOverflow
-#     [Documentation]    Category: social | URL: https://stackoverflow.com
-#     [Tags]    social    100websites
-    
-#     Open Website    https://stackoverflow.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=StackOverflow.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test GitHub_Explore
-#     [Documentation]    Category: social | URL: https://github.com/explore
-#     [Tags]    social    100websites
-    
-#     Open Website    https://github.com/explore
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=GitHub_Explore.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test GitHub_Trending
-#     [Documentation]    Category: social | URL: https://github.com/trending
-#     [Tags]    social    100websites
-    
-#     Open Website    https://github.com/trending
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=GitHub_Trending.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test ProductHunt
-#     [Documentation]    Category: social | URL: https://www.producthunt.com
-#     [Tags]    social    100websites
-    
-#     Open Website    https://www.producthunt.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=ProductHunt.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Lobsters
-#     [Documentation]    Category: social | URL: https://lobste.rs
-#     [Tags]    social    100websites
-    
-#     Open Website    https://lobste.rs
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Lobsters.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Slashdot
-#     [Documentation]    Category: social | URL: https://slashdot.org
-#     [Tags]    social    100websites
-    
-#     Open Website    https://slashdot.org
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Slashdot.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test IndieHackers
-#     [Documentation]    Category: social | URL: https://www.indiehackers.com
-#     [Tags]    social    100websites
-    
-#     Open Website    https://www.indiehackers.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=IndieHackers.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Amazon
-#     [Documentation]    Category: ecommerce | URL: https://www.amazon.com
-#     [Tags]    ecommerce    100websites
-    
-#     Open Website    https://www.amazon.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Amazon.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test eBay
-#     [Documentation]    Category: ecommerce | URL: https://www.ebay.com
-#     [Tags]    ecommerce    100websites
-    
-#     Open Website    https://www.ebay.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=eBay.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Etsy
-#     [Documentation]    Category: ecommerce | URL: https://www.etsy.com
-#     [Tags]    ecommerce    100websites
-    
-#     Open Website    https://www.etsy.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Etsy.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test AliExpress
-#     [Documentation]    Category: ecommerce | URL: https://www.aliexpress.com
-#     [Tags]    ecommerce    100websites
-    
-#     Open Website    https://www.aliexpress.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=AliExpress.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
-
-# Test Shopify
-#     [Documentation]    Category: ecommerce | URL: https://www.shopify.com
-#     [Tags]    ecommerce    100websites
-    
-#     Open Website    https://www.shopify.com
-    
-#     ${elements}=    Connector.Collect UI Candidates
-#     Log Many    @{elements}
-    
-#     ${screenshot}=    Take Screenshot    filename=Shopify.png    fullPage=True
-#     ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-#     [Teardown]    Close Browser
diff --git a/tests/atest/compounent/test_web_ui_js.robot b/tests/atest/compounent/test_web_ui_js.robot
deleted file mode 100644
index 7a038db..0000000
--- a/tests/atest/compounent/test_web_ui_js.robot
+++ /dev/null
@@ -1,1398 +0,0 @@
-*** Settings ***
-Documentation    Test JSQuery collector on 100 diverse websites
-...              This file is AUTO-GENERATED by generate_tests.py
-...              Categories: demo(20), simple(20), text(15), images(15), spa(15), social(10), ecommerce(5)
-Library          Browser
-Library          Agent.platforms._webconnector.WebConnectorRF    js_query    AS    Connector
-Library          tests/atest/compounent/BBoxVisualizer.py
-Resource         ${EXECDIR}/tests/atest/browsers/config.robot
-
-Suite Setup      Log    Starting test suite on 100 websites
-Suite Teardown   Log    ✅ Test suite complete
-
-*** Test Cases ***
-Test HerokuApp_Login
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/login
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/login
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Login.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test SauceDemo
-    [Documentation]    Category: demo | URL: https://www.saucedemo.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://www.saucedemo.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=SauceDemo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test DemoQA
-    [Documentation]    Category: demo | URL: https://demoqa.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://demoqa.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=DemoQA.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Dropdown
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/dropdown
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/dropdown
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Dropdown.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Checkboxes
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/checkboxes
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/checkboxes
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Checkboxes.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Inputs
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/inputs
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/inputs
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Inputs.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Upload
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/upload
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/upload
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Upload.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_Tables
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/tables
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/tables
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_Tables.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_IFrame
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/iframe
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/iframe
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_IFrame.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HerokuApp_DynamicLoading
-    [Documentation]    Category: demo | URL: https://the-internet.herokuapp.com/dynamic_loading/1
-    [Tags]    demo    100websites
-    
-    Open Website    https://the-internet.herokuapp.com/dynamic_loading/1
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HerokuApp_DynamicLoading.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test AutomationExercise
-    [Documentation]    Category: demo | URL: https://automationexercise.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://automationexercise.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=AutomationExercise.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test XYZ_Bank
-    [Documentation]    Category: demo | URL: https://www.globalsqa.com/angularJs-protractor/BankingProject
-    [Tags]    demo    100websites
-    
-    Open Website    https://www.globalsqa.com/angularJs-protractor/BankingProject
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=XYZ_Bank.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test PHPTravels_Demo
-    [Documentation]    Category: demo | URL: https://phptravels.com/demo
-    [Tags]    demo    100websites
-    
-    Open Website    https://phptravels.com/demo
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=PHPTravels_Demo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test OpenCart_Demo
-    [Documentation]    Category: demo | URL: https://demo.opencart.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://demo.opencart.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=OpenCart_Demo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test NopCommerce_Demo
-    [Documentation]    Category: demo | URL: https://demo.nopcommerce.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://demo.nopcommerce.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=NopCommerce_Demo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Magento_Demo
-    [Documentation]    Category: demo | URL: https://magento.softwaretestingboard.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://magento.softwaretestingboard.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Magento_Demo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test ExpandTesting_Practice
-    [Documentation]    Category: demo | URL: https://practice.expandtesting.com
-    [Tags]    demo    100websites
-    
-    Open Website    https://practice.expandtesting.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=ExpandTesting_Practice.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test EvilTester_Pages
-    [Documentation]    Category: demo | URL: https://testpages.eviltester.com/styled/index.html
-    [Tags]    demo    100websites
-    
-    Open Website    https://testpages.eviltester.com/styled/index.html
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=EvilTester_Pages.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test AutomationCamp
-    [Documentation]    Category: demo | URL: https://play1.automationcamp.ir/index.html
-    [Tags]    demo    100websites
-    
-    Open Website    https://play1.automationcamp.ir/index.html
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=AutomationCamp.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Example
-    [Documentation]    Category: simple | URL: https://example.com
-    [Tags]    simple    100websites
-    
-    Open Website    https://example.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Example.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test CERN_FirstWebsite
-    [Documentation]    Category: simple | URL: https://info.cern.ch
-    [Tags]    simple    100websites
-    
-    Open Website    https://info.cern.ch
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=CERN_FirstWebsite.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test W3C
-    [Documentation]    Category: simple | URL: https://www.w3.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.w3.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=W3C.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test IANA
-    [Documentation]    Category: simple | URL: https://www.iana.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.iana.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=IANA.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test RFC_Editor
-    [Documentation]    Category: simple | URL: https://www.rfc-editor.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.rfc-editor.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=RFC_Editor.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Kernel_Org
-    [Documentation]    Category: simple | URL: https://www.kernel.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.kernel.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Kernel_Org.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Debian
-    [Documentation]    Category: simple | URL: https://www.debian.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.debian.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Debian.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Python_Org
-    [Documentation]    Category: simple | URL: https://www.python.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.python.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Python_Org.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test NPM
-    [Documentation]    Category: simple | URL: https://www.npmjs.com
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.npmjs.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=NPM.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test PyPI
-    [Documentation]    Category: simple | URL: https://pypi.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://pypi.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=PyPI.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Rust_Lang
-    [Documentation]    Category: simple | URL: https://www.rust-lang.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.rust-lang.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Rust_Lang.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Go_Lang
-    [Documentation]    Category: simple | URL: https://golang.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://golang.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Go_Lang.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Ruby_Lang
-    [Documentation]    Category: simple | URL: https://www.ruby-lang.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.ruby-lang.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Ruby_Lang.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test PHP_Net
-    [Documentation]    Category: simple | URL: https://www.php.net
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.php.net
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=PHP_Net.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Perl_Org
-    [Documentation]    Category: simple | URL: https://www.perl.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.perl.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Perl_Org.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Scala_Lang
-    [Documentation]    Category: simple | URL: https://www.scala-lang.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.scala-lang.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Scala_Lang.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Kotlin_Lang
-    [Documentation]    Category: simple | URL: https://kotlinlang.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://kotlinlang.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Kotlin_Lang.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test SQLite
-    [Documentation]    Category: simple | URL: https://www.sqlite.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.sqlite.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=SQLite.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test PostgreSQL
-    [Documentation]    Category: simple | URL: https://www.postgresql.org
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.postgresql.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=PostgreSQL.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Nginx
-    [Documentation]    Category: simple | URL: https://www.nginx.com
-    [Tags]    simple    100websites
-    
-    Open Website    https://www.nginx.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Nginx.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_Python
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Python_(programming_language)
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Python_(programming_language)
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_Python.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_AI
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Artificial_intelligence
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Artificial_intelligence
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_AI.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_JavaScript
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/JavaScript
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/JavaScript
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_JavaScript.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_WebDev
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Web_development
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Web_development
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_WebDev.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_Testing
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Software_testing
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Software_testing
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_Testing.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_ML
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Machine_learning
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Machine_learning
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_ML.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_DataScience
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Data_science
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Data_science
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_DataScience.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_Cloud
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Cloud_computing
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Cloud_computing
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_Cloud.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_Blockchain
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Blockchain
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Blockchain
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_Blockchain.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Wikipedia_Cybersecurity
-    [Documentation]    Category: text | URL: https://en.wikipedia.org/wiki/Cybersecurity
-    [Tags]    text    100websites
-    
-    Open Website    https://en.wikipedia.org/wiki/Cybersecurity
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Wikipedia_Cybersecurity.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test BBC_News
-    [Documentation]    Category: text | URL: https://www.bbc.com/news
-    [Tags]    text    100websites
-    
-    Open Website    https://www.bbc.com/news
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=BBC_News.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Guardian_Tech
-    [Documentation]    Category: text | URL: https://www.theguardian.com/technology
-    [Tags]    text    100websites
-    
-    Open Website    https://www.theguardian.com/technology
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Guardian_Tech.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test DevTo
-    [Documentation]    Category: text | URL: https://dev.to
-    [Tags]    text    100websites
-    
-    Open Website    https://dev.to
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=DevTo.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test FreeCodeCamp_News
-    [Documentation]    Category: text | URL: https://www.freecodecamp.org/news
-    [Tags]    text    100websites
-    
-    Open Website    https://www.freecodecamp.org/news
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=FreeCodeCamp_News.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Smashing_Magazine
-    [Documentation]    Category: text | URL: https://www.smashingmagazine.com
-    [Tags]    text    100websites
-    
-    Open Website    https://www.smashingmagazine.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Smashing_Magazine.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Unsplash
-    [Documentation]    Category: images | URL: https://unsplash.com
-    [Tags]    images    100websites
-    
-    Open Website    https://unsplash.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Unsplash.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Pexels
-    [Documentation]    Category: images | URL: https://www.pexels.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.pexels.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Pexels.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Pixabay
-    [Documentation]    Category: images | URL: https://pixabay.com
-    [Tags]    images    100websites
-    
-    Open Website    https://pixabay.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Pixabay.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Flickr_Explore
-    [Documentation]    Category: images | URL: https://www.flickr.com/explore
-    [Tags]    images    100websites
-    
-    Open Website    https://www.flickr.com/explore
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Flickr_Explore.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test 500px
-    [Documentation]    Category: images | URL: https://500px.com
-    [Tags]    images    100websites
-    
-    Open Website    https://500px.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=500px.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Behance
-    [Documentation]    Category: images | URL: https://www.behance.net
-    [Tags]    images    100websites
-    
-    Open Website    https://www.behance.net
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Behance.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Dribbble
-    [Documentation]    Category: images | URL: https://dribbble.com
-    [Tags]    images    100websites
-    
-    Open Website    https://dribbble.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Dribbble.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Awwwards
-    [Documentation]    Category: images | URL: https://www.awwwards.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.awwwards.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Awwwards.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test DeviantArt
-    [Documentation]    Category: images | URL: https://www.deviantart.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.deviantart.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=DeviantArt.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test ArtStation
-    [Documentation]    Category: images | URL: https://www.artstation.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.artstation.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=ArtStation.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Shutterstock
-    [Documentation]    Category: images | URL: https://www.shutterstock.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.shutterstock.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Shutterstock.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Adobe_Stock
-    [Documentation]    Category: images | URL: https://stock.adobe.com
-    [Tags]    images    100websites
-    
-    Open Website    https://stock.adobe.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Adobe_Stock.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Getty_Images
-    [Documentation]    Category: images | URL: https://www.gettyimages.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.gettyimages.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Getty_Images.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test iStock
-    [Documentation]    Category: images | URL: https://www.istockphoto.com
-    [Tags]    images    100websites
-    
-    Open Website    https://www.istockphoto.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=iStock.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Burst_Shopify
-    [Documentation]    Category: images | URL: https://burst.shopify.com
-    [Tags]    images    100websites
-    
-    Open Website    https://burst.shopify.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Burst_Shopify.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test React_ShoppingCart
-    [Documentation]    Category: spa | URL: https://react-shopping-cart-67954.firebaseapp.com
-    [Tags]    spa    100websites
-    
-    Open Website    https://react-shopping-cart-67954.firebaseapp.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=React_ShoppingCart.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test TodoMVC_React
-    [Documentation]    Category: spa | URL: https://todomvc.com/examples/react
-    [Tags]    spa    100websites
-    
-    Open Website    https://todomvc.com/examples/react
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=TodoMVC_React.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test TodoMVC_Vue
-    [Documentation]    Category: spa | URL: https://todomvc.com/examples/vue
-    [Tags]    spa    100websites
-    
-    Open Website    https://todomvc.com/examples/vue
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=TodoMVC_Vue.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test TodoMVC_Angular
-    [Documentation]    Category: spa | URL: https://todomvc.com/examples/angular
-    [Tags]    spa    100websites
-    
-    Open Website    https://todomvc.com/examples/angular
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=TodoMVC_Angular.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test React_Homepage
-    [Documentation]    Category: spa | URL: https://reactjs.org
-    [Tags]    spa    100websites
-    
-    Open Website    https://reactjs.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=React_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Vue_Homepage
-    [Documentation]    Category: spa | URL: https://vuejs.org
-    [Tags]    spa    100websites
-    
-    Open Website    https://vuejs.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Vue_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Angular_Homepage
-    [Documentation]    Category: spa | URL: https://angular.io
-    [Tags]    spa    100websites
-    
-    Open Website    https://angular.io
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Angular_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Svelte_Homepage
-    [Documentation]    Category: spa | URL: https://svelte.dev
-    [Tags]    spa    100websites
-    
-    Open Website    https://svelte.dev
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Svelte_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test NextJS_Homepage
-    [Documentation]    Category: spa | URL: https://nextjs.org
-    [Tags]    spa    100websites
-    
-    Open Website    https://nextjs.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=NextJS_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test NuxtJS_Homepage
-    [Documentation]    Category: spa | URL: https://nuxtjs.org
-    [Tags]    spa    100websites
-    
-    Open Website    https://nuxtjs.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=NuxtJS_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Gatsby_Homepage
-    [Documentation]    Category: spa | URL: https://www.gatsbyjs.com
-    [Tags]    spa    100websites
-    
-    Open Website    https://www.gatsbyjs.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Gatsby_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Remix_Homepage
-    [Documentation]    Category: spa | URL: https://remix.run
-    [Tags]    spa    100websites
-    
-    Open Website    https://remix.run
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Remix_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Astro_Homepage
-    [Documentation]    Category: spa | URL: https://astro.build
-    [Tags]    spa    100websites
-    
-    Open Website    https://astro.build
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Astro_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Qwik_Homepage
-    [Documentation]    Category: spa | URL: https://qwik.builder.io
-    [Tags]    spa    100websites
-    
-    Open Website    https://qwik.builder.io
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Qwik_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Solid_Homepage
-    [Documentation]    Category: spa | URL: https://solidjs.com
-    [Tags]    spa    100websites
-    
-    Open Website    https://solidjs.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Solid_Homepage.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test HackerNews
-    [Documentation]    Category: social | URL: https://news.ycombinator.com
-    [Tags]    social    100websites
-    
-    Open Website    https://news.ycombinator.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=HackerNews.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Reddit_Programming
-    [Documentation]    Category: social | URL: https://www.reddit.com/r/programming
-    [Tags]    social    100websites
-    
-    Open Website    https://www.reddit.com/r/programming
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Reddit_Programming.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Reddit_WebDev
-    [Documentation]    Category: social | URL: https://www.reddit.com/r/webdev
-    [Tags]    social    100websites
-    
-    Open Website    https://www.reddit.com/r/webdev
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Reddit_WebDev.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test StackOverflow
-    [Documentation]    Category: social | URL: https://stackoverflow.com
-    [Tags]    social    100websites
-    
-    Open Website    https://stackoverflow.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=StackOverflow.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test GitHub_Explore
-    [Documentation]    Category: social | URL: https://github.com/explore
-    [Tags]    social    100websites
-    
-    Open Website    https://github.com/explore
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=GitHub_Explore.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test GitHub_Trending
-    [Documentation]    Category: social | URL: https://github.com/trending
-    [Tags]    social    100websites
-    
-    Open Website    https://github.com/trending
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=GitHub_Trending.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test ProductHunt
-    [Documentation]    Category: social | URL: https://www.producthunt.com
-    [Tags]    social    100websites
-    
-    Open Website    https://www.producthunt.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=ProductHunt.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Lobsters
-    [Documentation]    Category: social | URL: https://lobste.rs
-    [Tags]    social    100websites
-    
-    Open Website    https://lobste.rs
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Lobsters.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Slashdot
-    [Documentation]    Category: social | URL: https://slashdot.org
-    [Tags]    social    100websites
-    
-    Open Website    https://slashdot.org
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Slashdot.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test IndieHackers
-    [Documentation]    Category: social | URL: https://www.indiehackers.com
-    [Tags]    social    100websites
-    
-    Open Website    https://www.indiehackers.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=IndieHackers.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Amazon
-    [Documentation]    Category: ecommerce | URL: https://www.amazon.com
-    [Tags]    ecommerce    100websites
-    
-    Open Website    https://www.amazon.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Amazon.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test eBay
-    [Documentation]    Category: ecommerce | URL: https://www.ebay.com
-    [Tags]    ecommerce    100websites
-    
-    Open Website    https://www.ebay.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=eBay.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Etsy
-    [Documentation]    Category: ecommerce | URL: https://www.etsy.com
-    [Tags]    ecommerce    100websites
-    
-    Open Website    https://www.etsy.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Etsy.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test AliExpress
-    [Documentation]    Category: ecommerce | URL: https://www.aliexpress.com
-    [Tags]    ecommerce    100websites
-    
-    Open Website    https://www.aliexpress.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=AliExpress.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
-
-Test Shopify
-    [Documentation]    Category: ecommerce | URL: https://www.shopify.com
-    [Tags]    ecommerce    100websites
-    
-    Open Website    https://www.shopify.com
-    
-    ${elements}=    Connector.Collect UI Candidates
-    Log Many    @{elements}
-    
-    ${screenshot}=    Take Screenshot    filename=Shopify.png    fullPage=True
-    ${annotated}=    Draw And Log Elements    ${elements}    ${screenshot}
-    
-    [Teardown]    Close Browser
diff --git a/tests/atest/compounent/websites_100.yaml b/tests/atest/compounent/websites_100.yaml
deleted file mode 100644
index 84ca76f..0000000
--- a/tests/atest/compounent/websites_100.yaml
+++ /dev/null
@@ -1,413 +0,0 @@
-websites:
-  # ========== Sites Démo (20) ==========
-  - url: "https://the-internet.herokuapp.com/login"
-    name: "HerokuApp_Login"
-    category: "demo"
-  
-  - url: "https://www.saucedemo.com"
-    name: "SauceDemo"
-    category: "demo"
-  
-  - url: "https://demoqa.com"
-    name: "DemoQA"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/dropdown"
-    name: "HerokuApp_Dropdown"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/checkboxes"
-    name: "HerokuApp_Checkboxes"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/inputs"
-    name: "HerokuApp_Inputs"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/upload"
-    name: "HerokuApp_Upload"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/tables"
-    name: "HerokuApp_Tables"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/iframe"
-    name: "HerokuApp_IFrame"
-    category: "demo"
-  
-  - url: "https://the-internet.herokuapp.com/dynamic_loading/1"
-    name: "HerokuApp_DynamicLoading"
-    category: "demo"
-  
-  - url: "https://automationexercise.com"
-    name: "AutomationExercise"
-    category: "demo"
-  
-  - url: "https://www.globalsqa.com/angularJs-protractor/BankingProject"
-    name: "XYZ_Bank"
-    category: "demo"
-  
-  - url: "https://opensource-demo.orangehrmlive.com"
-    name: "OrangeHRM_Demo"
-    category: "demo"
-  
-  - url: "https://phptravels.com/demo"
-    name: "PHPTravels_Demo"
-    category: "demo"
-  
-  - url: "https://demo.opencart.com"
-    name: "OpenCart_Demo"
-    category: "demo"
-  
-  - url: "https://demo.nopcommerce.com"
-    name: "NopCommerce_Demo"
-    category: "demo"
-  
-  - url: "https://magento.softwaretestingboard.com"
-    name: "Magento_Demo"
-    category: "demo"
-  
-  - url: "https://practice.expandtesting.com"
-    name: "ExpandTesting_Practice"
-    category: "demo"
-  
-  - url: "https://testpages.eviltester.com/styled/index.html"
-    name: "EvilTester_Pages"
-    category: "demo"
-  
-  - url: "https://play1.automationcamp.ir/index.html"
-    name: "AutomationCamp"
-    category: "demo"
-
-  # ========== Sites Production Simples (20) ==========
-  - url: "https://example.com"
-    name: "Example"
-    category: "simple"
-  
-  - url: "https://info.cern.ch"
-    name: "CERN_FirstWebsite"
-    category: "simple"
-  
-  - url: "https://www.w3.org"
-    name: "W3C"
-    category: "simple"
-  
-  - url: "https://www.iana.org"
-    name: "IANA"
-    category: "simple"
-  
-  - url: "https://www.rfc-editor.org"
-    name: "RFC_Editor"
-    category: "simple"
-  
-  - url: "https://www.kernel.org"
-    name: "Kernel_Org"
-    category: "simple"
-  
-  - url: "https://www.debian.org"
-    name: "Debian"
-    category: "simple"
-  
-  - url: "https://www.python.org"
-    name: "Python_Org"
-    category: "simple"
-  
-  - url: "https://www.npmjs.com"
-    name: "NPM"
-    category: "simple"
-  
-  - url: "https://pypi.org"
-    name: "PyPI"
-    category: "simple"
-  
-  - url: "https://www.rust-lang.org"
-    name: "Rust_Lang"
-    category: "simple"
-  
-  - url: "https://golang.org"
-    name: "Go_Lang"
-    category: "simple"
-  
-  - url: "https://www.ruby-lang.org"
-    name: "Ruby_Lang"
-    category: "simple"
-  
-  - url: "https://www.php.net"
-    name: "PHP_Net"
-    category: "simple"
-  
-  - url: "https://www.perl.org"
-    name: "Perl_Org"
-    category: "simple"
-  
-  - url: "https://www.scala-lang.org"
-    name: "Scala_Lang"
-    category: "simple"
-  
-  - url: "https://kotlinlang.org"
-    name: "Kotlin_Lang"
-    category: "simple"
-  
-  - url: "https://www.sqlite.org"
-    name: "SQLite"
-    category: "simple"
-  
-  - url: "https://www.postgresql.org"
-    name: "PostgreSQL"
-    category: "simple"
-  
-  - url: "https://www.nginx.com"
-    name: "Nginx"
-    category: "simple"
-
-  # ========== Sites Riches en Texte (15) ==========
-  - url: "https://en.wikipedia.org/wiki/Python_(programming_language)"
-    name: "Wikipedia_Python"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Artificial_intelligence"
-    name: "Wikipedia_AI"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/JavaScript"
-    name: "Wikipedia_JavaScript"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Web_development"
-    name: "Wikipedia_WebDev"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Software_testing"
-    name: "Wikipedia_Testing"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Machine_learning"
-    name: "Wikipedia_ML"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Data_science"
-    name: "Wikipedia_DataScience"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Cloud_computing"
-    name: "Wikipedia_Cloud"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Blockchain"
-    name: "Wikipedia_Blockchain"
-    category: "text"
-  
-  - url: "https://en.wikipedia.org/wiki/Cybersecurity"
-    name: "Wikipedia_Cybersecurity"
-    category: "text"
-  
-  - url: "https://www.bbc.com/news"
-    name: "BBC_News"
-    category: "text"
-  
-  - url: "https://www.theguardian.com/technology"
-    name: "Guardian_Tech"
-    category: "text"
-  
-  - url: "https://dev.to"
-    name: "DevTo"
-    category: "text"
-  
-  - url: "https://www.freecodecamp.org/news"
-    name: "FreeCodeCamp_News"
-    category: "text"
-  
-  - url: "https://www.smashingmagazine.com"
-    name: "Smashing_Magazine"
-    category: "text"
-
-  # ========== Sites Riches en Images (15) ==========
-  - url: "https://unsplash.com"
-    name: "Unsplash"
-    category: "images"
-  
-  - url: "https://www.pexels.com"
-    name: "Pexels"
-    category: "images"
-  
-  - url: "https://pixabay.com"
-    name: "Pixabay"
-    category: "images"
-  
-  - url: "https://www.flickr.com/explore"
-    name: "Flickr_Explore"
-    category: "images"
-  
-  - url: "https://500px.com"
-    name: "500px"
-    category: "images"
-  
-  - url: "https://www.behance.net"
-    name: "Behance"
-    category: "images"
-  
-  - url: "https://dribbble.com"
-    name: "Dribbble"
-    category: "images"
-  
-  - url: "https://www.awwwards.com"
-    name: "Awwwards"
-    category: "images"
-  
-  - url: "https://www.deviantart.com"
-    name: "DeviantArt"
-    category: "images"
-  
-  - url: "https://www.artstation.com"
-    name: "ArtStation"
-    category: "images"
-  
-  - url: "https://www.shutterstock.com"
-    name: "Shutterstock"
-    category: "images"
-  
-  - url: "https://stock.adobe.com"
-    name: "Adobe_Stock"
-    category: "images"
-  
-  - url: "https://www.gettyimages.com"
-    name: "Getty_Images"
-    category: "images"
-  
-  - url: "https://www.istockphoto.com"
-    name: "iStock"
-    category: "images"
-  
-  - url: "https://burst.shopify.com"
-    name: "Burst_Shopify"
-    category: "images"
-
-  # ========== Applications SPA/React (15) ==========
-  - url: "https://react-shopping-cart-67954.firebaseapp.com"
-    name: "React_ShoppingCart"
-    category: "spa"
-  
-  - url: "https://todomvc.com/examples/react"
-    name: "TodoMVC_React"
-    category: "spa"
-  
-  - url: "https://todomvc.com/examples/vue"
-    name: "TodoMVC_Vue"
-    category: "spa"
-  
-  - url: "https://todomvc.com/examples/angular"
-    name: "TodoMVC_Angular"
-    category: "spa"
-  
-  - url: "https://reactjs.org"
-    name: "React_Homepage"
-    category: "spa"
-  
-  - url: "https://vuejs.org"
-    name: "Vue_Homepage"
-    category: "spa"
-  
-  - url: "https://angular.io"
-    name: "Angular_Homepage"
-    category: "spa"
-  
-  - url: "https://svelte.dev"
-    name: "Svelte_Homepage"
-    category: "spa"
-  
-  - url: "https://nextjs.org"
-    name: "NextJS_Homepage"
-    category: "spa"
-  
-  - url: "https://nuxtjs.org"
-    name: "NuxtJS_Homepage"
-    category: "spa"
-  
-  - url: "https://www.gatsbyjs.com"
-    name: "Gatsby_Homepage"
-    category: "spa"
-  
-  - url: "https://remix.run"
-    name: "Remix_Homepage"
-    category: "spa"
-  
-  - url: "https://astro.build"
-    name: "Astro_Homepage"
-    category: "spa"
-  
-  - url: "https://qwik.builder.io"
-    name: "Qwik_Homepage"
-    category: "spa"
-  
-  - url: "https://solidjs.com"
-    name: "Solid_Homepage"
-    category: "spa"
-
-  # ========== Réseaux Sociaux / Forums (10) ==========
-  - url: "https://news.ycombinator.com"
-    name: "HackerNews"
-    category: "social"
-  
-  - url: "https://www.reddit.com/r/programming"
-    name: "Reddit_Programming"
-    category: "social"
-  
-  - url: "https://www.reddit.com/r/webdev"
-    name: "Reddit_WebDev"
-    category: "social"
-  
-  - url: "https://stackoverflow.com"
-    name: "StackOverflow"
-    category: "social"
-  
-  - url: "https://github.com/explore"
-    name: "GitHub_Explore"
-    category: "social"
-  
-  - url: "https://github.com/trending"
-    name: "GitHub_Trending"
-    category: "social"
-  
-  - url: "https://www.producthunt.com"
-    name: "ProductHunt"
-    category: "social"
-  
-  - url: "https://lobste.rs"
-    name: "Lobsters"
-    category: "social"
-  
-  - url: "https://slashdot.org"
-    name: "Slashdot"
-    category: "social"
-  
-  - url: "https://www.indiehackers.com"
-    name: "IndieHackers"
-    category: "social"
-
-  # ========== E-commerce (5) ==========
-  - url: "https://www.amazon.com"
-    name: "Amazon"
-    category: "ecommerce"
-  
-  - url: "https://www.ebay.com"
-    name: "eBay"
-    category: "ecommerce"
-  
-  - url: "https://www.etsy.com"
-    name: "Etsy"
-    category: "ecommerce"
-  
-  - url: "https://www.aliexpress.com"
-    name: "AliExpress"
-    category: "ecommerce"
-  
-  - url: "https://www.shopify.com"
-    name: "Shopify"
-    category: "ecommerce"
-
-
-
-
-
-
diff --git a/tests/atest/devices/_browserstack.robot b/tests/atest/config/_browserstack.robot
similarity index 85%
rename from tests/atest/devices/_browserstack.robot
rename to tests/atest/config/_browserstack.robot
index dff81fa..ce4988e 100644
--- a/tests/atest/devices/_browserstack.robot
+++ b/tests/atest/config/_browserstack.robot
@@ -1,6 +1,6 @@
 *** Settings ***
 Documentation    Keywords for BrowserStack testing
-Variables    ${EXECDIR}/tests/atest/devices/settings.yaml
+Variables    ${EXECDIR}/tests/atest/config/settings.yaml
 Variables    ${EXECDIR}/browserstack.yml
 
 *** Variables ***
diff --git a/tests/atest/devices/_local.robot b/tests/atest/config/_local.robot
similarity index 96%
rename from tests/atest/devices/_local.robot
rename to tests/atest/config/_local.robot
index c15149e..c7ce5e1 100644
--- a/tests/atest/devices/_local.robot
+++ b/tests/atest/config/_local.robot
@@ -1,6 +1,6 @@
 *** Settings ***
 Documentation    Keywords for local testing
-Variables    ${EXECDIR}/tests/atest/devices/settings.yaml
+Variables    ${EXECDIR}/tests/atest/config/settings.yaml
 
 *** Variables ***
 # Variables calculées depuis settings.yaml
diff --git a/tests/atest/config/browsers.robot b/tests/atest/config/browsers.robot
new file mode 100644
index 0000000..62f9b4b
--- /dev/null
+++ b/tests/atest/config/browsers.robot
@@ -0,0 +1,29 @@
+*** Settings ***
+Documentation    Web browser configuration
+Variables    ${EXECDIR}/tests/atest/config/settings.yaml
+Library    Browser
+Library    Agent    llm_client=openai    llm_model=gpt-4o
+
+*** Keywords ***
+Open Website
+    [Documentation]    Opens browser and optionally navigates to URL
+    ...                Args: url (optional) - URL to navigate to
+    ...                element_source: tree | visual (default: tree)
+    ...                selection_mode: text | som (default: text)
+    [Arguments]    ${url}=${EMPTY}   ${element_source}=dom    ${selection_mode}=som
+    Setup Agent Mode        ${element_source}     ${selection_mode}
+    New Browser    ${browser_type}    headless=${headless_mode}    
+    New Context    locale=en-US
+    Log    Browser opened - Type: ${browser_type}, Headless: ${headless_mode}
+    New Page    ${url}
+    Set Viewport Size    width=1920    height=1080
+
+
+Setup Agent Mode
+    [Documentation]    Configure Agent element source and selection modes
+    [Arguments]    ${element_source}    ${selection_mode}
+    ${agent}=    Get Library Instance    Agent
+    Call Method    ${agent.engine}    set_element_source    ${element_source}
+    Call Method    ${agent.engine}    set_llm_input_format    ${selection_mode}
+    Log    Agent configured - element_source: ${element_source}, selection_mode: ${selection_mode}
+
diff --git a/tests/atest/config/devices.robot b/tests/atest/config/devices.robot
new file mode 100644
index 0000000..d679f0d
--- /dev/null
+++ b/tests/atest/config/devices.robot
@@ -0,0 +1,42 @@
+*** Settings ***
+Documentation    Mobile device configuration for Local and BrowserStack
+Variables    ${EXECDIR}/tests/atest/config/settings.yaml
+Library    AppiumLibrary
+Library    Agent    llm_client=openai    llm_model=gpt-4o   platform_type=mobile
+
+*** Keywords ***
+Open Application With Config
+    [Documentation]    Opens application based on Device_location setting
+    ...                Options: Local or Browserstack
+    ...                element_source: tree | visual (default: tree)
+    [Arguments]    ${element_source}=dom    ${selection_mode}=som
+    Setup Agent Mode    ${element_source}    ${selection_mode}
+    Run Keyword If    '${Device_location}' == 'Local'    Open Local Application
+    ...    ELSE IF    '${Device_location}' == 'Browserstack'    Open BrowserStack Application
+    ...    ELSE    Fail    Device_location invalid: ${Device_location}
+
+Open Local Application
+    [Documentation]    Opens application with local configuration
+    Import Resource    ${EXECDIR}/tests/atest/config/_local.robot
+    Run Keyword If    '${platformName}' == 'Android'    Open Application    remote_url=${remote_url}    &{DESIRED_CAPABILITIES_Android}
+    ...    ELSE IF    '${platformName}' == 'iOS'    Open Application    remote_url=${remote_url}    &{DESIRED_CAPABILITIES_iOS}
+    ...    ELSE    Fail    platformName invalid: ${platformName}
+    Log    Local application opened - Platform: ${platformName}
+
+Open BrowserStack Application
+    [Documentation]    Opens application with BrowserStack configuration
+    
+    Import Variables    ${EXECDIR}/browserstack.yml
+    Import Resource    ${EXECDIR}/tests/atest/config/_browserstack.robot
+    Open Application    remote_url=${remote_url}
+    Log    BrowserStack application opened
+
+Setup Agent Mode
+    [Documentation]    Configure Agent element source and selection modes
+    [Arguments]    ${element_source}    ${selection_mode}
+    ${agent}=    Get Library Instance    Agent
+    Call Method    ${agent.engine}    set_element_source    ${element_source}
+    Call Method    ${agent.engine}    set_llm_input_format    ${selection_mode}
+    Log    Agent configured - element_source: ${element_source}, selection_mode: ${selection_mode}
+
+
diff --git a/tests/atest/devices/config.robot b/tests/atest/devices/config.robot
deleted file mode 100644
index fc86679..0000000
--- a/tests/atest/devices/config.robot
+++ /dev/null
@@ -1,29 +0,0 @@
-*** Settings ***
-Documentation    Configuration centralisée pour Local et BrowserStack
-Variables    ${EXECDIR}/tests/atest/devices/settings.yaml
-Library    AppiumLibrary
-
-*** Keywords ***
-Open Application With Config
-    [Documentation]    Ouvre l'application selon Device_location dans settings.yaml
-    ...                Options: Local ou Browserstack
-    Run Keyword If    '${Device_location}' == 'Local'    Open Local Application
-    ...    ELSE IF    '${Device_location}' == 'Browserstack'    Open BrowserStack Application
-    ...    ELSE    Fail    Device_location invalide: ${Device_location}
-
-Open Local Application
-    [Documentation]    Ouvre l'application avec configuration locale
-    Import Resource    ${EXECDIR}/tests/atest/devices/_local.robot
-    Run Keyword If    '${platformName}' == 'Android'    Open Application    remote_url=${remote_url}    &{DESIRED_CAPABILITIES_Android}
-    ...    ELSE IF    '${platformName}' == 'iOS'    Open Application    remote_url=${remote_url}    &{DESIRED_CAPABILITIES_iOS}
-    ...    ELSE    Fail    platformName invalide: ${platformName}
-    Log    Application locale ouverte - Platform: ${platformName}
-
-Open BrowserStack Application
-    [Documentation]    Ouvre l'application avec configuration BrowserStack (via BrowserStack SDK)
-    Import Variables    ${EXECDIR}/browserstack.yml
-    Import Resource    ${EXECDIR}/tests/atest/devices/_browserstack.robot
-    # BrowserStack SDK gère automatiquement les capabilities depuis browserstack.yml
-    Open Application    remote_url=${remote_url}
-    Log    Application BrowserStack ouverte
-
diff --git a/tests/atest/devices/settings.yaml b/tests/atest/devices/settings.yaml
deleted file mode 100644
index 4d29205..0000000
--- a/tests/atest/devices/settings.yaml
+++ /dev/null
@@ -1,61 +0,0 @@
-# ============================================================================
-# DEVICE CONFIGURATION
-# ============================================================================
-# See README.md in this folder for detailed documentation
-#
-# Quick Start:
-# 1. Set Device_location: Local or Browserstack
-# 2. Configure your device in the appropriate section below
-# 3. Run: robot tests/atest/e2e/your_test.robot
-#
-# ============================================================================
-# Global Settings
-# ============================================================================
-Device_location: Local  # Options: Local | Browserstack
-platformName: Android
-
-# Android App Configuration
-app_package_Android: com.skill2lead.appiumdemo
-app_activity_Android: com.skill2lead.appiumdemo.MainActivity
-
-# Wait strategy
-delay: 1
-veryShort_wait: 3
-short_wait: 7
-normal_wait: 13
-long_wait: 25
-veryLong_wait: 40
-
-# ============================================================================
-# Local Device Settings
-# ============================================================================
-# Local Appium Server
-remote_url_local: http://127.0.0.1:4723
-
-# iOS Local Device Details
-deviceName_iOS: iPhone rouge
-platformName_iOS: iOS
-platformVersion_iOS: 15.2.1
-udid: 00008030-0012351234567890
-bundleId: com.vianavigo.iphone
-appPath_iOS:  
-xcodeOrgId: 
-xcodeSigningId: 
-automationName_iOS: XCUITest
-AllowPopupCap_iOS: autoAcceptAlerts
-language_iOS: fr
-locale_iOS: fr
-fullReset_iOS: false
-
-# Android Local Device Details
-deviceName_Android: 192.168.50.235:5555
-platformName_Android: Android
-platformVersion_Android: 13.0
-# Android local device Settings
-automationName_Android: Uiautomator2
-AllowPopupCap_Android: autoGrantPermissions
-language_Android: fr
-locale_Android: fr
-noReset_Android: false
-
-
diff --git a/tests/atest/e2e/atest_web_ask.robot b/tests/atest/keywords/web/ask_scenario.robot
similarity index 92%
rename from tests/atest/e2e/atest_web_ask.robot
rename to tests/atest/keywords/web/ask_scenario.robot
index b4431bc..51f1058 100644
--- a/tests/atest/e2e/atest_web_ask.robot
+++ b/tests/atest/keywords/web/ask_scenario.robot
@@ -1,6 +1,6 @@
 *** Settings ***
 Documentation    Tests for Agent.Ask keyword
-Resource    ../browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 Test Setup    Open Website    https://agent-arena-eta.vercel.app/v1/
 Test Teardown    Close Browser
 
diff --git a/tests/atest/e2e/atest_web_find_visual.robot b/tests/atest/keywords/web/find_visual.robot
similarity index 93%
rename from tests/atest/e2e/atest_web_find_visual.robot
rename to tests/atest/keywords/web/find_visual.robot
index 2ed6d6b..f323b0d 100644
--- a/tests/atest/e2e/atest_web_find_visual.robot
+++ b/tests/atest/keywords/web/find_visual.robot
@@ -1,6 +1,6 @@
 *** Settings ***
 Documentation    Tests for Agent.Find Visual Element keyword
-Resource    ../browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 Test Setup    Open Website    https://agent-arena-eta.vercel.app/snapshots/v1/arena/search/c2/
 Test Teardown    Close Browser
 
diff --git a/tests/atest/e2e/atest_mobile_h_datepicker.robot b/tests/atest/tools/mobile/test_datepicker.robot
similarity index 97%
rename from tests/atest/e2e/atest_mobile_h_datepicker.robot
rename to tests/atest/tools/mobile/test_datepicker.robot
index 990ecad..5a22076 100644
--- a/tests/atest/e2e/atest_mobile_h_datepicker.robot
+++ b/tests/atest/tools/mobile/test_datepicker.robot
@@ -3,7 +3,7 @@ Documentation    E2E tests for Date picker - Happy paths
 Library    AppiumLibrary
 Library    Agent
 Library    DateTime
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Keywords ***
 Get Current Date Plus Days
diff --git a/tests/atest/e2e/atest_mobile_f_longclick.robot b/tests/atest/tools/mobile/test_gestures.robot
similarity index 94%
rename from tests/atest/e2e/atest_mobile_f_longclick.robot
rename to tests/atest/tools/mobile/test_gestures.robot
index 7dd9144..d9a6453 100644
--- a/tests/atest/e2e/atest_mobile_f_longclick.robot
+++ b/tests/atest/tools/mobile/test_gestures.robot
@@ -2,7 +2,7 @@
 Documentation    E2E tests for advanced interactions - Happy paths
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test Long Click
diff --git a/tests/atest/e2e/atest_mobile_a_inputvalue.robot b/tests/atest/tools/mobile/test_input.robot
similarity index 92%
rename from tests/atest/e2e/atest_mobile_a_inputvalue.robot
rename to tests/atest/tools/mobile/test_input.robot
index 5506e4f..ebe1578 100644
--- a/tests/atest/e2e/atest_mobile_a_inputvalue.robot
+++ b/tests/atest/tools/mobile/test_input.robot
@@ -2,7 +2,7 @@
 Documentation    E2E test for basic form interaction - Happy path
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test Basic Form Input And Submit
diff --git a/tests/atest/e2e/atest_mobile_c_scrollview.robot b/tests/atest/tools/mobile/test_scrollview.robot
similarity index 96%
rename from tests/atest/e2e/atest_mobile_c_scrollview.robot
rename to tests/atest/tools/mobile/test_scrollview.robot
index 664bfed..cb65e0b 100644
--- a/tests/atest/e2e/atest_mobile_c_scrollview.robot
+++ b/tests/atest/tools/mobile/test_scrollview.robot
@@ -2,7 +2,7 @@
 Documentation    E2E test for ScrollView - Happy path
 Library    AppiumLibrary
 Library    Agent
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Test Cases ***
 Test ScrollView Scroll Positif - Correct Buttons Display
diff --git a/tests/atest/e2e/atest_mobile_g_timepicker.robot b/tests/atest/tools/mobile/test_timepicker.robot
similarity index 98%
rename from tests/atest/e2e/atest_mobile_g_timepicker.robot
rename to tests/atest/tools/mobile/test_timepicker.robot
index 1f57230..baea507 100644
--- a/tests/atest/e2e/atest_mobile_g_timepicker.robot
+++ b/tests/atest/tools/mobile/test_timepicker.robot
@@ -3,7 +3,7 @@ Documentation    E2E tests for Time picker - Happy paths
 Library    AppiumLibrary
 Library    Agent
 Library    DateTime
-Resource    ${EXECDIR}/tests/atest/devices/config.robot
+Resource    ${EXECDIR}/tests/atest/config/devices.robot
 
 *** Keywords ***
 Get Device Time In 12H Format
diff --git a/tests/atest/e2e/atest_web_e_navigation.robot b/tests/atest/tools/web/test_back.robot
similarity index 90%
rename from tests/atest/e2e/atest_web_e_navigation.robot
rename to tests/atest/tools/web/test_back.robot
index 3d9dc04..1c53b02 100644
--- a/tests/atest/e2e/atest_web_e_navigation.robot
+++ b/tests/atest/tools/web/test_back.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for browser navigation using go_back tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Go Back Navigation
diff --git a/tests/atest/e2e/atest_web_vision_basic.robot b/tests/atest/tools/web/test_basic.robot
similarity index 80%
rename from tests/atest/e2e/atest_web_vision_basic.robot
rename to tests/atest/tools/web/test_basic.robot
index fd0175b..9c7eb33 100644
--- a/tests/atest/e2e/atest_web_vision_basic.robot
+++ b/tests/atest/tools/web/test_basic.robot
@@ -1,7 +1,6 @@
 *** Settings ***
 Documentation    Simple vision mode tests for basic web interactions
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
-Suite Setup    Open Website
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 Suite Teardown    Close Browser
 
 *** Test Cases ***
@@ -9,7 +8,7 @@ Test Click Simple Button
     [Documentation]    Test clicking a simple button using vision mode
     [Tags]    vision    web    click    basic
     
-    Go To    https://the-internet.herokuapp.com/
+    Open Website    https://the-internet.herokuapp.com/    visual    som
     Sleep    ${short_wait}
     
     Agent.Do    click on Add/Remove Elements link
@@ -21,7 +20,7 @@ Test Click And Verify
     [Documentation]    Test clicking and verifying result using vision mode
     [Tags]    vision    web    click    verify
     
-    Go To    https://the-internet.herokuapp.com/add_remove_elements/
+    Open Website    https://the-internet.herokuapp.com/add_remove_elements/    visual    som
     Sleep    ${short_wait}
     
     Agent.Do    click on Add Element button
@@ -34,7 +33,7 @@ Test Hover Over Element
     [Documentation]    Test hovering over an element using vision mode
     [Tags]    vision    web    hover
     
-    Go To    https://the-internet.herokuapp.com/hovers
+    Open Website    https://the-internet.herokuapp.com/hovers    visual    som
     Sleep    ${short_wait}
     
     Agent.Do    hover over the first image
@@ -47,7 +46,7 @@ Test Navigate Back
     [Documentation]    Test navigation using vision mode
     [Tags]    vision    web    navigation
     
-    Go To    https://the-internet.herokuapp.com/
+    Open Website    https://the-internet.herokuapp.com/       visual    som
     Sleep    ${short_wait}
     
     Agent.Do    click on Checkboxes link
diff --git a/tests/atest/e2e/atest_web_h_clear_text.robot b/tests/atest/tools/web/test_clear.robot
similarity index 92%
rename from tests/atest/e2e/atest_web_h_clear_text.robot
rename to tests/atest/tools/web/test_clear.robot
index 66ee19e..186129e 100644
--- a/tests/atest/e2e/atest_web_h_clear_text.robot
+++ b/tests/atest/tools/web/test_clear.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for clear text tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Clear Text From Input Field
diff --git a/tests/atest/e2e/atest_web_g_double_click.robot b/tests/atest/tools/web/test_click.robot
similarity index 90%
rename from tests/atest/e2e/atest_web_g_double_click.robot
rename to tests/atest/tools/web/test_click.robot
index 1f1aff9..9b7148d 100644
--- a/tests/atest/e2e/atest_web_g_double_click.robot
+++ b/tests/atest/tools/web/test_click.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for double click tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Double Click On Element
diff --git a/tests/atest/e2e/atest_web_f_hover.robot b/tests/atest/tools/web/test_hover.robot
similarity index 90%
rename from tests/atest/e2e/atest_web_f_hover.robot
rename to tests/atest/tools/web/test_hover.robot
index 95a8151..b7a6b26 100644
--- a/tests/atest/e2e/atest_web_f_hover.robot
+++ b/tests/atest/tools/web/test_hover.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for hover tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Hover Over Element
diff --git a/tests/atest/e2e/atest_web_a_basic_form.robot b/tests/atest/tools/web/test_input.robot
similarity index 92%
rename from tests/atest/e2e/atest_web_a_basic_form.robot
rename to tests/atest/tools/web/test_input.robot
index 3cf8cc9..217f582 100644
--- a/tests/atest/e2e/atest_web_a_basic_form.robot
+++ b/tests/atest/tools/web/test_input.robot
@@ -1,6 +1,6 @@
 *** Settings ***
 Documentation    E2E test for basic web form interaction - Happy path
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Basic Form Input And Submit
diff --git a/tests/atest/e2e/atest_web_c_press_key.robot b/tests/atest/tools/web/test_press_key.robot
similarity index 91%
rename from tests/atest/e2e/atest_web_c_press_key.robot
rename to tests/atest/tools/web/test_press_key.robot
index c39dfdc..432d544 100644
--- a/tests/atest/e2e/atest_web_c_press_key.robot
+++ b/tests/atest/tools/web/test_press_key.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for keyboard key press using press_key tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Press Enter Key To Submit
diff --git a/tests/atest/e2e/atest_web_d_scroll.robot b/tests/atest/tools/web/test_scroll.robot
similarity index 92%
rename from tests/atest/e2e/atest_web_d_scroll.robot
rename to tests/atest/tools/web/test_scroll.robot
index 0ef92ea..2a822a2 100644
--- a/tests/atest/e2e/atest_web_d_scroll.robot
+++ b/tests/atest/tools/web/test_scroll.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for page scrolling using scroll_down tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Scroll Down Page
diff --git a/tests/atest/e2e/atest_web_j_scroll_to_element.robot b/tests/atest/tools/web/test_scroll_to_element.robot
similarity index 90%
rename from tests/atest/e2e/atest_web_j_scroll_to_element.robot
rename to tests/atest/tools/web/test_scroll_to_element.robot
index e9ae26b..af0f35a 100644
--- a/tests/atest/e2e/atest_web_j_scroll_to_element.robot
+++ b/tests/atest/tools/web/test_scroll_to_element.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for scroll to element tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Scroll To Specific Element
diff --git a/tests/atest/e2e/atest_web_i_scroll_up.robot b/tests/atest/tools/web/test_scroll_up.robot
similarity index 91%
rename from tests/atest/e2e/atest_web_i_scroll_up.robot
rename to tests/atest/tools/web/test_scroll_up.robot
index dbac60c..3473566 100644
--- a/tests/atest/e2e/atest_web_i_scroll_up.robot
+++ b/tests/atest/tools/web/test_scroll_up.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for scroll up tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Scroll Up Page
diff --git a/tests/atest/e2e/atest_web_b_select_dropdown.robot b/tests/atest/tools/web/test_select.robot
similarity index 92%
rename from tests/atest/e2e/atest_web_b_select_dropdown.robot
rename to tests/atest/tools/web/test_select.robot
index 6ef9852..ee02e22 100644
--- a/tests/atest/e2e/atest_web_b_select_dropdown.robot
+++ b/tests/atest/tools/web/test_select.robot
@@ -1,7 +1,7 @@
 *** Settings ***
 Documentation    E2E test for dropdown selection using select_option tool
 Library    Browser
-Resource    ${EXECDIR}/tests/atest/browsers/config.robot
+Resource    ${EXECDIR}/tests/atest/config/browsers.robot
 
 *** Test Cases ***
 Test Select Option From Dropdown
diff --git a/tests/utest/platforms/test_js_collector.py b/tests/utest/platforms/test_js_collector.py
index 91059d2..b1c14f3 100644
--- a/tests/utest/platforms/test_js_collector.py
+++ b/tests/utest/platforms/test_js_collector.py
@@ -1,4 +1,7 @@
 import pytest
+
+pytest.skip("Web support coming soon - js_query_collector not yet implemented", allow_module_level=True)
+
 from Agent.platforms.collectors.js_query_collector import JSQueryCollector
 
 
diff --git a/tests/utest/platforms/test_web_locator.py b/tests/utest/platforms/test_web_locator.py
index 8993cf9..697534c 100644
--- a/tests/utest/platforms/test_web_locator.py
+++ b/tests/utest/platforms/test_web_locator.py
@@ -1,4 +1,7 @@
 import pytest
+
+pytest.skip("Web support coming soon - web locator not yet implemented", allow_module_level=True)
+
 from Agent.platforms.locators.web import WebLocatorBuilder