ksachdeva/repo-stargazer

Implements semantic search over starred repository README files using naive RAG (chunking + embeddings stored in a vector database), then feeds results to an AI agent for interactive exploration. Supports multiple embedding providers (Ollama, OpenAI, Azure OpenAI) via LiteLLM and exposes functionality through Google ADK's web interface. Downloads repository metadata and READMEs via GitHub API, builds a local vector store in the user's data directory, and requires a GitHub Personal Access Token to avoid rate limits.

No commits in the last 6 months. Available on PyPI.