Knowledge Agent

A powerful CLI tool for understanding codebases through natural language queries, featuring role-based explanations and advanced code analysis capabilities. Built with LangChain and Claude.

Key Features

• 🤖 Intelligent Code Understanding

  • Natural language queries about your codebase
  • Semantic code search and analysis
  • Multi-language support with tree-sitter
  • Advanced code parsing and documentation generation

• 🎭 Role-Based Interactions

  • Customizable explanation styles for different roles
  • Dynamic role switching during sessions
  • Persistent conversation memory
  • Context-aware responses

• 🔍 Advanced Search & Analysis

  • Semantic code search with ChromaDB
  • Code structure analysis
  • Documentation parsing (Markdown, RST, PDF)
  • Version control integration

• 📚 Documentation Features

  • Automated documentation generation
  • Architecture diagram creation
  • MkDocs integration with Material theme
  • API documentation

Installation

1. Clone the repository:

git clone https://github.com/Gajesh2007/knowledge-agent.git
cd knowledge-agent

2. Install using Poetry (Python 3.9+):

# Install Poetry if needed
curl -sSL https://install.python-poetry.org | python3 -

# Install dependencies
poetry install

3. Configure environment:

cp .env.example .env
# Edit .env with required settings:
# - ANTHROPIC_API_KEY: Your Claude API key
# - Other optional configurations

Quick Start

1. Ingest your codebase:

# Local codebase
knowledge-agent ingest --path ./your/code

# From GitHub
knowledge-agent fetch --repo https://github.com/user/repo --ingest

2. Query your codebase:

# One-off query
knowledge-agent search "How does the error handling work?"

# Interactive session
knowledge-agent session

3. Generate documentation:

knowledge-agent docs generate ./src

Advanced Usage

Role-Based Queries

# Specify role for targeted explanations
knowledge-agent search --role engineer "How is authentication implemented?"
knowledge-agent search --role beginner "Explain the basic architecture"

Interactive Sessions

# Start session with specific role
knowledge-agent session --role architect

# Available commands in session:
# - /role <role>: Switch roles
# - /clear: Clear conversation history
# - /exit: End session

Documentation Generation

# Generate with specific template
knowledge-agent docs generate ./src --template api

# Include architecture diagram
knowledge-agent docs generate ./src --with-diagram

# Custom output
knowledge-agent docs generate ./src --output ./custom-docs

Version Control Integration

# Analyze specific version
knowledge-agent version analyze --repo ./repo --ref v1.0.0

# Compare versions
knowledge-agent version diff --repo ./repo --old v1.0.0 --new v2.0.0

Project Structure

knowledge-agent/
├── knowledge_agent/
│   ├── core/                # Core functionality
│   │   ├── llm.py          # LLM integration
│   │   ├── retrieval.py    # Search functionality
│   │   ├── code_parser.py  # Code analysis
│   │   ├── doc_parser.py   # Documentation parsing
│   │   └── ...
│   └── cli/                # CLI interface
├── tests/                  # Test suite
└── docs/                   # Documentation

Development

# Install dev dependencies
poetry install --with dev

# Run tests
pytest
pytest --cov=knowledge_agent

# Code quality
black knowledge_agent tests
isort knowledge_agent tests
flake8 knowledge_agent tests
mypy knowledge_agent

Dependencies

• Python 3.9+
• Key packages:
  • LangChain 0.3.17
  • Anthropic Claude SDK 0.45.2
  • ChromaDB 0.6.3
  • Tree-sitter for code parsing
  • MkDocs with Material theme
  • Various documentation parsers

Contributing

1. Fork the repository
2. Create a feature branch
3. Make changes and run tests
4. Submit a pull request

Follow Conventional Commits for commit messages.

License

MIT License - See LICENSE file for details.