Skip to content

Its an autonomous Agent which will do testing and debugging, Test generation etc. This Project is created through Google AI Studio

Notifications You must be signed in to change notification settings

ksharma6309/HelixDevAgent_TestAndCodeAutomator_GoogleAIStudio

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

7 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

Helix DevAgent AI 🧬

Helix DevAgent AI is a futuristic, autonomous agentic suite designed to evolve the Software Development Life Cycle (SDLC).

Powered by Google Gemini 2.5 Flash, it serves as a neural interface where developers command specialized AI agents to handle unit testing, debugging, code auditing, and architectural strategy.

License React TypeScript Gemini


πŸ’‘ The Vision

The Problem

Modern software engineering involves a high "context-switching tax." Developers spend hours writing boilerplate tests, parsing obscure error logs, and debating style in code reviews. Existing tools are fragmented across CLIs, CI/CD pipelines, and web portals.

The Solution

Helix DevAgent AI unifies these workflows into a Single Pane of Glass. Instead of manually writing a Jest suite, you activate the Test Forge. Instead of Googling a stack trace, you deploy the Auto-Debugger. It transforms the IDE from a text editor into an Intelligent Command Center.


πŸš€ Modules

1. Helix Assistant views/ChatAssistant.tsx

  • Role: Central Intelligence & Strategy.
  • Capabilities: Answers high-level architectural questions, explains complex DevOps concepts, and retains context across the session in a persistent local database.
  • UI: High-contrast, sharp interface for maximum readability.

2. Project Explorer views/ProjectExplorer.tsx

  • Role: Codebase Navigation & Analysis.
  • Capabilities: Upload local files/folders. Features a floating "Summarize File" button for instant AI summaries of code logic. Allows selecting code snippets to "Explain" or "Implement" new features directly via the Assistant.

3. Test Forge views/TestGenerator.tsx

  • Input: Raw source code (Python, JS, Go, Java, etc.).
  • Output: Comprehensive unit test suites (Jest, PyTest, JUnit).
  • Behavior: Covers happy paths, edge cases, and mocks dependencies automatically.

4. Auto-Debugger views/Debugger.tsx

  • Input: Broken code + Error logs/Stack traces.
  • Output: Root cause analysis and fixed code snippets.
  • Behavior: Correlates logic errors with runtime exceptions to provide a precise fix.

5. Code Auditor views/CodeReview.tsx

  • Input: Code snippets.
  • Output: Detailed Markdown report covering Security (OWASP), Performance, and Clean Code principles.
  • Persistence: Automatically saves review history for compliance auditing.

6. Log Sentinel views/LogAnalyzer.tsx

  • Input: Unstructured raw server logs.
  • Output: Structured incident reports with severity classification and remediation steps.

7. Refactor Engine views/RefactorBot.tsx

  • Input: Legacy or "Spaghetti" code.
  • Output: Modernized, optimized, and readable code reducing cyclomatic complexity.

8. Data Vault views/DatabaseManager.tsx

  • Privacy First: All data is stored locally in the browser (LocalStorage).
  • Portability: Export your entire session history to a JSON database file (helix_db.json) and import it on any machine.

πŸ“‚ Feature Spotlight: Project Explorer

The Project Explorer is the neural hub for interacting with your local codebase. It provides a file-system interface directly within the browser, augmented by AI context actions.

Core Capabilities

1. Summarize File

  • Trigger: Click the "Summarize" button in the toolbar or the floating action button that appears on hover.
  • Function: Sends the entire file content to the Helix Assistant.
  • Output: A concise, bullet-point executive summary of the file. It highlights:
    • Primary Responsibility: What the file does.
    • Key Functions: Important methods and their purposes.
    • Dependencies: Critical imports and external connections.

2. Implement Feature

  • Trigger: Click the "Implement" button.
  • Function: Opens a dialogue with the Assistant pre-loaded with the file's context.
  • Workflow: You describe a new feature (e.g., "Add a dark mode toggle" or "Handle API error retries"). The AI generates the specific code modifications required to implement that feature within the existing file structure.

3. Explain Code & Selection

  • Full File Explanation: The "Explain" button provides a comprehensive walkthrough of the entire file's logic, data flow, and architecture.
  • Smart Selection: Highlight any snippet of text within the code viewer. A "Explain Selection" button dynamically appears. Clicking this focuses the AI on just that specific block, perfect for decoding complex algorithms or regex patterns without losing context of the surrounding code.

⚑ Installation & Setup

Helix is designed as a modern React application. While it can run in containerized AI environments, you can also run it locally.

Prerequisites

Setup Steps

  1. Clone the Repository

    git clone https://github.com/ksharma6309/helix-devagent.git
    cd helix-devagent
  2. Install Dependencies (If using a bundler like Vite/Create React App)

    npm install
  3. Configure Environment Create a .env file in the root directory:

    REACT_APP_API_KEY=your_actual_gemini_api_key

    Note: The application expects process.env.API_KEY to be available.

  4. Run the Application

    npm start
    # or
    npm run dev
  5. Access Helix Open your browser to http://localhost:3000 (or the port specified by your runner).


🧠 Why Google AI Studio?

We chose Google AI Studio and the Gemini 2.5 Flash model as the core engine for Helix because of its specific advantages for real-time developer tools.

For the Developer

  • Low Latency & High Throughput: Gemini 2.5 Flash is optimized for speed, making real-time code generation and debugging feel instantaneous.
  • Large Context Window: It can ingest massive log files or long code snippets (up to 1M tokens) without losing context.
  • Multimodal Capabilities: It can natively understand text, code, and structured data seamlessly.

πŸ› οΈ Tech Stack


πŸ“‚ Project Structure

/
β”œβ”€β”€ index.html              # Entry point
β”œβ”€β”€ src/
β”‚   β”œβ”€β”€ App.tsx             # Main Router & Layout
β”‚   β”œβ”€β”€ components/         # Reusable UI (Buttons, Sidebar, Markdown)
β”‚   β”œβ”€β”€ services/           # AI Integration (Gemini Service)
β”‚   β”œβ”€β”€ utils/              # Storage & Persistence Logic
β”‚   β”œβ”€β”€ views/              # Main Application Screens
β”‚   β”‚   β”œβ”€β”€ Dashboard.tsx       # Command Center
β”‚   β”‚   β”œβ”€β”€ ChatAssistant.tsx   # Helix Assistant
β”‚   β”‚   β”œβ”€β”€ ProjectExplorer.tsx # File Navigation & Summarization
β”‚   β”‚   β”œβ”€β”€ TestGenerator.tsx   # Test Forge
β”‚   β”‚   β”œβ”€β”€ Debugger.tsx        # Auto-Debugger
β”‚   β”‚   β”œβ”€β”€ CodeReview.tsx      # Code Auditor
β”‚   β”‚   └── DatabaseManager.tsx # Data Vault
β”‚   └── types.ts            # TypeScript Interfaces
└── README.md

πŸ“· Screenshots

Dashboard - Screenshot 2025-12-15 160711

Chatbot Assistant - Screenshot 2025-12-15 162938

Project Explorer - image

Test Generator - Screenshot 2025-12-15 175413

Code Debugger - Screenshot 2025-12-15 175615

Code Reviewer - Screenshot 2025-12-15 175711

Data Logs - Screenshot 2025-12-15 175737


πŸ“ License

This project is licensed under the MIT License - see the LICENSE file for details.


πŸ‘©β€πŸ’» Author

Khushboo Sharma

GitHub: https://github.com/ksharma6309

LinkedIn: https://www.linkedin.com/in/khushboo-sharma-b5b372125/

About

Its an autonomous Agent which will do testing and debugging, Test generation etc. This Project is created through Google AI Studio

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published