Skip to content

REFACTOR.md

Latest commit

 

History

History
82 lines (69 loc) · 3.3 KB

REFACTOR.md

File metadata and controls

82 lines (69 loc) · 3.3 KB

Source Registry & Architecture

This project archives everything Anthropic publishes that helps people build with Claude: docs, engineering posts, research, SDK references, cookbooks, skills, plugins, and support articles.

Sources (2,900+ files, 11 sources)

Source Section Pages Method
code.claude.com claude-code 141 llms.txt -> .md suffix
platform.claude.com api 1,541 sitemap -> .md suffix
modelcontextprotocol.io mcp 203 sitemap -> .md suffix
anthropic.com/engineering engineering 26 jina.ai proxy
anthropic.com/research research 118 jina.ai proxy
anthropic.com/news (filtered) news ~76 jina.ai proxy
anthropic.com/product blog 4 jina.ai proxy
support.claude.com support 343 jina.ai proxy
github.com/anthropics/* github 718 raw.githubusercontent.com
NPM manifest meta 1 registry.npmjs.org
GitHub CHANGELOG meta 1 raw.githubusercontent.com

Discovery

fetcher.py --discover probes all known Anthropic domains via robots.txt, sitemap.xml, llms.txt, and llms-full.txt. Run periodically to find new sources.

Known domains: anthropic.com, platform.claude.com, code.claude.com, support.claude.com, modelcontextprotocol.io, claude.ai, claude.com

Content Layout

content/
  en/docs/claude-code/   Claude Code + Agent SDK (code.claude.com)
  en/api/                API reference (platform.claude.com)
  en/build-with-claude/  Platform features
  en/agents-and-tools/   Tool use, agent skills
  en/manage-claude/      Admin, billing
  en/managed-agents/     Managed agents API
  mcp/                   MCP protocol spec + community
  blog/
    engineering/         Technical deep-dives
    research/            Research papers
    news/                Model releases, announcements
    product/             Product pages
  support/               Help articles (support.claude.com)
  github/
    cookbooks/           claude-cookbooks (recipes, notebooks)
    skills/              Official Agent Skills
    plugins-official/    Plugin directory
    courses/             Prompt engineering curriculum
    quickstarts/         Deployable app starters
    code-action/         GitHub Actions for Claude Code
    workshops/           CWC workshops
    long-running-agents/ Long-running agent patterns
    sdk-python/          Python SDK docs
    sdk-typescript/      TypeScript SDK docs

Commands

uv run scripts/fetcher.py                         # Fetch everything
uv run scripts/fetcher.py --section mcp            # MCP spec only
uv run scripts/fetcher.py --section github         # GitHub repos only
uv run scripts/fetcher.py --section engineering    # Engineering blog
uv run scripts/fetcher.py --incremental            # Skip existing
uv run scripts/fetcher.py --tree                   # Show all sources
uv run scripts/fetcher.py --discover               # Probe for new sources
uv run scripts/fetcher.py URL [URL ...]            # Fetch specific URLs

GitHub repo fetching needs GITHUB_TOKEN or GH_TOKEN env var. Blog/research/support use jina.ai which rate-limits at ~10 req/s.

Source Registry

sources.json contains the complete machine-readable registry of all discovered Anthropic content sources, including domains, aliases, and per-source metadata.