This example demonstrates how to use Codegen to build a CLI tool for deep code research. The tool allows you to:

• Clone and analyze any GitHub repository
• Ask questions about the codebase
• Explore dependencies and relationships
• Search for patterns and implementations

1. Install the requirements:

uv venv
source .venv/bin/activate
uv sync

1. Set up your OpenAI API key in a .env:

OPENAI_API_KEY=your-api-key

Run the CLI tool by providing a GitHub repository:

python run.py research "owner/repo"

For example:

python run.py research "fastapi/fastapi"

You can also provide an initial query:

python run.py research "fastapi/fastapi" -q "Explain the main components"

• "Explain the main components and their relationships"
• "Find all usages of the FastAPI class"
• "Show me the dependency graph for the routing module"
• "What design patterns are used in this codebase?"
• "How is dependency injection implemented?"

The research agent has access to several powerful tools:

• Semantic code search
• Symbol relationship analysis
• Directory structure exploration
• Code viewing and analysis

The agent maintains conversation history, so you can ask follow-up questions and build on previous findings.

Type "exit" or "quit" to end the research session.