"Roll for Initiative... in Python!"
This repository contains my code and progress through the AWS Strands SDK Workshop. It tracks my transformation from a coding apprentice into a master of AI agent orchestration, building a digital adventuring party capable of autonomous thought and collaboration.
The project is structured into five epic chapters, each building upon the last:
- Goal: Master the fundamental ritual of agent creation.
- Key Learning: Configuring system prompts and model parameters within the Strands framework.
- Goal: Equip agents with built-in magical tools.
- Key Learning: Implementing autonomous web scraping and information gathering with tool safety mechanisms.
- Goal: Create the legendary Dice of Destiny.
- Key Learning: Transforming Python functions into agent-executable tools using the
@tooldecorator.
- Goal: Connect to external realms via Model Context Protocol.
- Key Learning: Building MCP servers and clients to bridge agents with external services.
- Goal: Command a full D&D adventuring party.
- Key Learning: Orchestrating Agent-to-Agent (A2A) communication for complex, distributed AI tasks.
- Language: Python
- Framework: Strands SDK
- AI Models: Amazon Bedrock (Claude 3 / 3.5)
- Integration: Model Context Protocol (MCP)
- Clone the Quest Log
git clone https://github.com/noecrn/mcp-agent-framework.git
cd strands-quest- Prepare your Environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
pip install -r requirements.txt- Configure AWS Credentials Ensure your AWS CLI is configured with permissions for Amazon Bedrock.
- The Dice of Destiny: A custom tool allowing agents to simulate high-stakes RPG rolls.
- Multi-Agent Coordination: Watching a "Dungeon Master" agent delegate lore-checking to a "Librarian" agent.
“The best way to predict the future is to build the agents that will create it.”