A CLI that keeps one set of Markdown rules in .ruler/ and generates the instruction and MCP config files for over 30 AI coding agents.
eli64s/readme-ai
A CLI tool that generates README files for your repository using LLMs, or offline without an API key.
- Stars
- 3,005
- Forks
- 296
- Open issues
- 61
- Last commit
- 4 days ago
What is it?
ReadmeAI is a command-line tool that creates README files by analyzing a codebase. You point it at a GitHub, GitLab, or Bitbucket URL or a local directory, and it extracts project structure, dependencies, and usage details to produce a structured README.
It works with multiple LLM providers, including OpenAI, Anthropic, Google Gemini, and Ollama for local models. An offline mode generates README files without calling any LLM API. Output can be customized through options for header and navigation styles, badge colors and styles, logos, emoji packs, and directory tree depth. A .readmeaiignore file controls which files are analyzed.
The tool is distributed on PyPI and as a Docker image, and it can also be run from source. It is aimed at developers who want to bootstrap or refresh documentation for projects without writing it by hand.
Key features
- 01Generates README files from a GitHub, GitLab, Bitbucket URL or a local directory.
- 02Supports OpenAI, Anthropic, Google Gemini, and Ollama as LLM providers.
- 03Offers an offline mode that needs no LLM API service.
- 04Customizes headers, navigation, badges, logos, and emojis via CLI options.
- 05Uses a .readmeaiignore file to exclude files from analysis.
Use cases
- Bootstrapping a README for a new project that lacks documentation.
- Refreshing outdated README files across several repositories.
- Generating documentation for a private codebase using a local model via Ollama.
Quick start
- 1
Install ReadmeAI following the Getting Started section of the README
- 2
Generate a README for a repository by passing its URL
readmeai --repository https://github.com/eli64s/readme-ai
Things to know
- LLM-generated output requires an API key from a provider such as OpenAI, Anthropic, Gemini, or a local Ollama setup.
In collections
Topics
FAQ
- What is readme-ai?
- A CLI tool that generates README files for your repository using LLMs, or offline without an API key. ReadmeAI is a command-line tool that creates README files by analyzing a codebase. You point it at a GitHub, GitLab, or Bitbucket URL or a local directory, and it extracts project structure, dependencies, and usage details to produce a structured README.
- Is readme-ai free to use? What license does it have?
- readme-ai is open source under the MIT license. Check the license terms before using it in commercial products.
- Is readme-ai still maintained?
- Actively maintained. The last commit to readme-ai was on Nov 4, 2026.
- What can I use readme-ai for?
- Bootstrapping a README for a new project that lacks documentation. Refreshing outdated README files across several repositories. Generating documentation for a private codebase using a local model via Ollama.
- How do I install readme-ai?
- Install ReadmeAI following the Getting Started section of the README Generate a README for a repository by passing its URL: readmeai --repository https://github.com/eli64s/readme-ai
- Where can I get readme-ai?
- The source code and installation instructions are on GitHub: https://github.com/eli64s/readme-ai
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Generates end-to-end Playwright tests from your PRDs, ADRs and READMEs, and flags when the product and its documentation disagree.
A Markdown syntax reference and test file showing how GitHub Flavored Markdown (GFM) renders in READMEs, issues and wikis.
A CLI that writes and maintains a linked Markdown wiki for your codebase, grounded in source evidence, for your coding agents and you.
A local-first CLI and browser UI for reviewing AI coding agents' plans, git diffs and running apps with inline comments the agent can act on.
A plugin for Claude Code and Codex that writes end-user help articles from your source code, with screenshots rendered from your own UI markup.