What it does

ReadmeAI is a Python command-line tool that analyzes a repository, from a local path or a GitHub, GitLab or Bitbucket URL, and generates a README for it. The output can include an introduction, a features table, project structure and index, setup and usage guides, and contribution and community sections, with style options for badges, headers, navigation, logos and emojis. It is model agnostic. You can use OpenAI, Anthropic or Google Gemini with your own API key, run local models through Ollama with no key, or use an offline mode that builds a README without calling any language model. A .readmeaiignore file lets you exclude files from analysis. Install it with pip, pipx or uv, or run the Docker image. It needs Python 3.10 or later and is MIT licensed.

Capabilities

Generates a README from a local path or a GitHub, GitLab or Bitbucket repository
Switch between OpenAI, Anthropic, Gemini and Ollama
Offline mode that needs no LLM service
Style options for badges, headers, navigation, logos and emojis
.readmeaiignore patterns to control which files are analyzed
Install with pip, pipx, uv or Docker

When to use it

Starting documentation for a project that has no README yet
Keeping README structure consistent across many repositories
Generating documentation locally with Ollama or offline mode instead of a hosted API