Gitingest
Turns any Git repo into a single LLM-ready text digest by swapping 'hub' for 'ingest' in the URL.
Pick Gitingest if you want a zero-setup way to dump a GitHub repo into Claude or ChatGPT for review, onboarding, or refactor planning.
Skip it if you need persistent code search, embeddings, or an agent that can navigate a repo on its own.
Gitingest is a small utility that flattens a Git repository into a single prompt-friendly text digest: file tree plus concatenated source, with token counts and basic include/exclude filters. The headline trick is the URL rewrite: replace 'github.com' with 'gitingest.com' on any public repo and you get an instantly shareable, copy-pasteable bundle suitable for dropping into Claude, ChatGPT, Gemini, or any other long-context model.
It is aimed at developers doing code review, onboarding, refactor planning, or one-shot codebase Q&A with an LLM, where setting up a full retrieval pipeline is overkill. The hosted web version is free; the project itself is open source (MIT, coderamp-labs/gitingest on GitHub) and ships as a Python package and CLI for local use, plus a Chrome extension that adds an 'Ingest' button to GitHub. Private repos are supported via a personal access token which the site says is not persisted, and cloned repos are deleted after processing.
It is not a RAG system, agent, or hosted chat - it is a preprocessor. For very large monorepos you will still blow past context windows and need to scope with the include/exclude patterns, but as a glue tool between GitHub and a frontier LLM it is hard to beat.
Gitingest is the kind of two-feature tool that quietly becomes part of your daily workflow: change one word in a URL, get a clean codebase digest. The fact that it is MIT-licensed and self-hostable means you are not betting on the hosted service sticking around, which is exactly the right posture for a utility this small.
— The AI Tool Bible editorial team
Pros
- ✅ URL-swap trick (hub -> ingest) makes any public repo instantly LLM-ready
- ✅ Open source (MIT) with Python package, CLI, and Chrome extension
- ✅ Shows token counts and supports include/exclude filters out of the box
- ✅ Works with private repos via PAT that the service says it does not store
Cons
- ⚠️ Not a RAG or chat layer - you still paste output into an LLM yourself
- ⚠️ Large monorepos will exceed context windows without aggressive filtering
- ⚠️ Hosted instance has no formal SLA or paid tier for heavy use
Use cases
Explore related
Compare with similar tools
All in Coding →Cursor
FeaturedAI-first VS Code fork — chat, edit, and agentic coding in one IDE.
GitHub Copilot
FeaturedThe original AI pair programmer, now with chat and agents.
Replit Agent
FeaturedBuild & deploy a full app from a single prompt.
Aider
Terminal-based AI pair programmer that writes commits.
Codeium
Free, fast AI autocomplete + chat across 70+ editors.
Cody
Sourcegraph's AI coding assistant — codebase-aware via their search index.