Resource library

The living list from appendix A6. Every entry is fetched and verified before it earns a card; the full library ships with the appendices phase.

Claude Code documentationdocs
The source of truth this course verifies every Claude Code fact against.
Claude Code docs mapdocs
One fetchable index of every doc page. The staying-current habit starts here.
Claude Platform docsdocs
The Claude API reference: models, tool use, prompt caching, everything Part 3 builds on.
Claude Agent SDK overviewdocs
Where 2.11 and the harness lessons live. Read this before writing your first SDK agent.
Best practices for Claude Codeengineering
Anthropic's own operating patterns; Part 2 assumes you have read it.
Effective context engineering for AI agentsengineering
The canonical writeup behind lesson 1.3's context model.
Writing effective tools for AI agentsengineering
Tool design doctrine you will apply when building MCP servers and SDK agents.
Building effective agentsengineering
Workflows versus agents, and when not to build one. The pattern language Part 3 uses.
How we built our multi-agent research systemengineering
The honest account of orchestrator-worker tradeoffs behind the subagent lessons.
Contextual retrievalengineering
The retrieval technique appendix A8's RAG build implements. Read it before arguing about chunking.
Model Context Protocolprotocol
The protocol spec and guides behind lesson 2.7 and appendix A1.
MCP specification (latest)protocol
The versioned wire spec. Check here when a server and client disagree about capabilities.
anthropics/claude-coderepo
The official repo: issues and releases are where behavior changes surface first.
anthropics/skillsrepo
Official public Agent Skills; the reference for skill authoring in 2.4.
anthropics/claude-plugins-officialrepo
The official plugin catalog 2.8 tours. Read a plugin's source here before you install it.
anthropics/claude-code-actionrepo
Claude Code in GitHub Actions; the CI automation surface lesson 2.9 builds on.
modelcontextprotocol/serversrepo
Reference MCP servers maintained by the protocol team. Steal patterns, not just servers.
modelcontextprotocol/typescript-sdkrepo
The official TypeScript SDK you will use when appendix A1 has you build a server.
github/github-mcp-serverrepo
GitHub's official MCP server, the one 2.7 connects. First-party beats community wrappers.
microsoft/playwright-mcprepo
Browser automation over MCP; how your agent verifies UI work in the capstones.
getsentry/sentry-mcprepo
First-party Sentry MCP server; the production-debugging loop from the ops lessons.
pgvector/pgvectorrepo
Vector similarity search in Postgres; what A8's RAG pipeline stores embeddings in.
typescript-lsp plugintooling
Real language server intelligence for Claude. Highest value-per-risk in the catalog per 2.8.
commit-commands plugintooling
Git workflow skills from the official catalog; the install example lab 2.8 audits.
pr-review-toolkit plugintooling
Review agents for PR-shaped work; pairs with the reviewer patterns you build in 2.10.
security-guidance plugintooling
An extra security reviewer on every change. Soft enforcement, not a gate; treat it that way.
hookify plugintooling
Generates 2.6-style hooks from plain descriptions. Audit its output like any copied hook.
Supabase docscapstone
Auth, Postgres, and RLS for the capstones. The RLS guides are the part most people skip and regret.
Stripe docscapstone
Billing and webhooks for Track A. Verify event handling here, not from an agent's memory.
Vercel docscapstone
Deployment target for both capstones: cron jobs, function duration limits, environments.
Next.js docscapstone
The App Router reference the capstone builds assume. Agents hallucinate old APIs; check here.