AutomaDocs
automadocs.com
- Free plan
- Web
- API
- Self-hosted
What AutomaDocs is
AutomaDocs creates documentation from GitHub repositories and updates it when code changes are pushed. It can produce README files, API references, function notes, architecture overviews, and getting-started guides. Its AI chat answers plain-language questions about a codebase and points to relevant files and functions. JavaScript, TypeScript, and Python receive deep Tree-sitter parsing; Claude AI can analyze other languages directly. GitHub OAuth and webhooks connect repositories, and documentation can be used with Claude Code and Cursor through MCP. Users can edit generated pages, which AutomaDocs says it preserves and merges with later AI updates. Documentation health scores show the percentage of functions with documentation: 80% and above is green, while below 40% is red. The free plan covers one repository and 50 AI chat messages per month. Pro is $39/mo (annual). Managed cloud is the default, with Docker Compose and setup scripts available for self-hosting.
Who it suits
AutomaDocs describes its audience as engineering managers and tech leads at small and mid-size product teams that ship in GitHub. It may suit teams that want generated documentation alongside codebase questions and repository updates.
Good to know, and what to watch
- Generates several documentation types from GitHub repositories
- Refreshes documentation when code changes are pushed
- AI chat cites relevant files and functions
- Supports edits merged with later AI updates
- Self-hosting setup is available
- Free plan is limited to one repository
- Free plan includes 50 AI chat messages monthly
- No 24/7 phone line, SLA, or account manager
Plans
Every plan and priceCustom
Free
Pro
Team
Free plan
The verdict
AutomaDocs combines repository-linked documentation generation with codebase Q&A and automatic updates. Teams should weigh its free-plan limits and support model against their needs; Pro is $39/mo (annual).
What the maker publishes
- AI chat
- The chat answers plain-language repository questions using hybrid keyword and semantic retrieval and links to files and line numbers. source
- Authentication security
- Authentication uses GitHub OAuth, optional Google OAuth, JWTs, HTTP-only cookies, and CSRF protection. source
- Automatic synchronization
- GitHub webhooks trigger regeneration for changed files after pushes, typically completing in under two minutes. source
- Changelog update
- A March 2026 release fixed webhook auto-sync reliability and added weekly health-score email reports. source
- Code analysis
- Tree-sitter parsing combined with Claude analyzes code structure, signatures, relationships, and patterns. source
- Code questions
- Its AI chat answers plain-language questions about a codebase and cites the relevant files and functions. source
- Company founder
- AutomaDocs identifies Liz Stein as its founder and says she built it after repeatedly answering code questions in Slack. source
- Company headquarters
- No headquarters location is stated on the reviewed official AutomaDocs pages. source
- Compliance limitation
- AutomaDocs says it does not claim a specific third-party certification such as SOC 2 unless it publishes an attestation. source
- Data deletion
- The privacy policy says repository documentation and embeddings are removed within 24 hours after repository deletion, and account deletion removes associated data immediately. source
- Data handling
- The privacy policy says repository content is processed by Anthropic Claude, may be processed by OpenAI if Claude is temporarily unavailable, and is not used by subprocessors to train AI models. source
- Data processors
- Repository data is processed by Anthropic Claude, Pinecone, and sometimes OpenAI; the policy says subprocessors do not train models on customer code. source
- Deletion
- Repository documentation and embeddings are removed within 24 hours after repository deletion, while account deletion removes data immediately. source
- Developer surfaces
- AutomaDocs advertises a CLI, GitHub Action, MCP server, and integrations with Claude Code, Cursor, Windsurf, and VS Code. source
- Documentation types
- It can generate README files, API references, function notes, architecture overviews, and getting-started guides. source
- Documentation types
- It can generate READMEs, API references, function notes, architecture overviews, and getting-started guides. source
- Editing
- Users can edit generated pages, and the service says it preserves and merges their changes with AI updates. source
- Editing
- Users can edit generated documentation pages. source
- Export
- Documentation can be exported as Markdown. source
- Founded year
- No formal company founding year is stated; the changelog records the product's public launch in January 2026. source
- GitHub integration
- GitHub OAuth and webhooks are supported, with an optional GitHub App for enhanced permissions. source
- Health scores
- Documentation health scores report the percentage of functions with documentation; scores of 80% or higher are green and below 40% are red. source
- Health scores
- Health scoring tracks documentation coverage, freshness, and completeness and can send weekly reports. source
- Human review guidance
- The feature documentation recommends treating generated pages as drafts and reviewing examples, edge cases, and naming before publication. source
Checked against the maker's pages on 15 Oct 2026.