GitHub Repo Docs Scraper for RAG & AI Agents
Pricing
Pay per usage
GitHub Repo Docs Scraper for RAG & AI Agents
Fetch documentation from GitHub repositories — READMEs, docs folders, wikis — and convert to clean, chunked markdown optimized for RAG pipelines. Use via Apify Console/API or connect as an MCP server for Claude, Cursor, and other AI agents.
Pricing
Pay per usage
Rating
0.0
(0)
Developer
Aaron Hampton
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
0
Monthly active users
3 days ago
Last modified
Categories
Share
Fetch documentation from GitHub repositories — READMEs, docs folders, and custom paths — and convert to clean, chunked markdown optimized for RAG pipelines and AI agents.
Perfect for building RAG systems over your own repos, analyzing open-source documentation, and feeding code docs to AI agents.
Features
- README extraction — fetch and chunk any repo's README
- Docs folder traversal — recursively scan
docs/,documentation/,doc/, and custom paths - Markdown chunking — header-aware splitting with configurable chunk size and overlap
- Rich metadata — every chunk includes file path, line numbers, heading, and token count
- Optional GitHub token — supports authenticated requests (5000 req/hr vs 60 unauthenticated)
- Dual-mode — run as a normal Apify Actor OR connect as an MCP server for Claude, Cursor, and other AI agents
- No browser required — uses the GitHub REST API directly
Use Cases
- RAG pipelines — chunk repo docs and ingest into vector databases for Q&A over code documentation
- AI agents — let Claude/Cursor read repo documentation via MCP tools
- Documentation analysis — compare docs across repos, find gaps, track changes
- Onboarding — generate summaries of a repo's documentation for new contributors
- Search — keyword search across a repo's entire documentation
Input (Normal Actor Mode)
| Field | Type | Description |
|---|---|---|
repoUrl | string | GitHub repo URL or owner/repo shorthand |
githubToken | string | Optional PAT for higher rate limits |
includeReadme | boolean | Fetch README (default true) |
includeDocsFolder | boolean | Scan docs/ directory (default true) |
docsPaths | array | Additional paths to scan |
maxFiles | integer | Max files to fetch (default 50, max 500) |
chunkSize | integer | Words per chunk (default 1000) |
chunkOverlap | integer | Overlap between chunks (default 200) |
Output
{"repo": "facebook/react","filePath": "README.md","fileName": "README.md","fileType": "md","size": 5317,"url": "https://github.com/facebook/react/blob/main/README.md","contentMarkdown": "# React · ...","chunks": [{"id": "facebook_react_README_md_0","text": "# React ...","tokens": 465,"filePath": "README.md","startLine": 1,"endLine": 10,"heading": "React"}]}
MCP Tools
When connected as an MCP server, the following tools are available:
| Tool | Description |
|---|---|
get_readme | Fetch and chunk a repo's README. |
get_docs | Fetch and chunk files from a docs folder. |
search_repo_docs | Fetch docs and filter by keyword. |
Pricing
Pay-per-event. First results are free; subsequent events billed at:
| Event | Price |
|---|---|
| File fetched | $0.003 |
| Chunk generated | $0.001 |
| MCP tool call | $0.01 |
| Actor start | $0.00005 |
Volume discounts available at higher tiers.