AI Documentation Generator
Multi-agent documentation generator: analyzes a repository and builds a README and AI assistant config files
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- Reads the whole codebase and writes documentation files
- In GitLab mode it opens merge requests automatically
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/ai-documentation-generatorDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Run one by one in the Claude Code chat
/plugin marketplace add divar-ir/ai-doc-gen
/plugin install ai-doc-gen@divarInstall the plugin: /plugin marketplace add divar-ir/ai-doc-gen, then /plugin install ai-doc-gen@divar. Use the analyze-codebase and generate-readme skills.
Other ways from the author
/plugin marketplace add divar-ir/ai-doc-gen
/plugin install ai-doc-gen@divarNo keys or Python; skills analyze-codebase, generate-readme, generate-ai-rules.
This is third-party code. Review the repository files before installing.
What it does
AI Documentation Generator dissects a codebase and builds documentation from it with several agents. Five analysis agents concurrently map structure, dependencies, data flow, request flow and APIs, and generator agents turn the analysis into a README and AI assistant config files: CLAUDE.md, AGENTS.md and Cursor rules. It installs as a Claude Code plugin with the analyze-codebase, generate-readme and generate-ai-rules skills, or runs as a Python application. A GitLab mode discovers active projects and opens merge requests with fresh analysis on its own.
Who it is for. For technical writers and developers who need to document a repository fast.
Good fit when
- You need a README and a codebase map from a repository
- You need CLAUDE.md, AGENTS.md and Cursor rules
- You need to auto-update docs in GitLab via merge requests
Not a fit when
- You want hand-written docs with tight control over every paragraph
- You have no access to an OpenAI-compatible LLM for the Python mode
Example request
Analyze this repository and build a README with sections on structure, dependencies and APIsLimitations
The Claude Code plugin install needs no keys or Python. The Python mode needs Python 3.13, Git and access to an OpenAI-compatible LLM provider (OpenAI, compatible gateways, OpenRouter, local models). A Dockerfile and Helm chart are provided for containers and scheduled runs.
How to disable. Uninstall the plugin via /plugin uninstall ai-doc-gen@divar; for the Python mode, uninstall the package.
Security check
- Reads the whole codebase and writes documentation files
- In GitLab mode it opens merge requests automatically
README in short
The README describes AI Documentation Generator as a multi-agent code documentation generator. Five analysis agents concurrently map structure, dependencies, data and request flows and APIs and write reusable documents to .ai/docs/, and generators produce a README and CLAUDE.md, AGENTS.md and Cursor rules. It installs as a Claude Code plugin with skills that need no keys or Python, while the Python mode via uv or pip needs Python 3.13 and an OpenAI-compatible provider. A GitLab mode discovers active projects and opens merge requests with analysis, and a Dockerfile and Helm chart are included.
FAQ
Is an API key needed?
Not for the Claude Code plugin install; the skills run directly. The Python mode needs access to an OpenAI-compatible provider.
What does it generate besides a README?
AI assistant configs: CLAUDE.md, AGENTS.md and Cursor rules, with the analysis written to .ai/docs/.
Related
Microsoft's utility and MCP server that turn PDF, Word, Excel, PowerPoint, HTML and other files into Markdown for models
Anthropic skills
Skills
Anthropic's official skill examples: docx, pdf, pptx and xlsx documents, design, MCP builder and the Agent Skills spec
A skill that builds editable PPTX decks from PDF, DOCX, URLs and Markdown with native shapes, charts and animations
A skill that draws editorial diagrams as standalone HTML and SVG files: architecture, flowcharts, ER, funnels and about forty more types