Overview
Gitingest converts Git repositories into LLM-friendly text digests, available as a CLI, Python package, browser extensions, and self-hostable server. It produces compact summaries, token counts, and structured outputs optimized for prompts.
Core Features
- Generate text digests from repository URLs or local directories, including subdirectories and submodules.
- Provide statistics (file tree, size, token estimates) and optimized formatting for LLM consumption.
- Offer CLI, Python API, browser extensions, and optional server deployment for ingestion pipelines.
Use Cases
- Preparing high-quality text slices for RAG pipelines, code search, or code Q&A assistants.
- Automating codebase audits, migration summaries, and contextual prompt building for LLMs.
Technical Highlights
- Python-first implementation with async interfaces, supports pip/pipx installation and Docker-based deployments.
- Configurable output formats, token estimation, and sharding strategies for large repositories to reduce model costs.