Anthropic's official "Code with Claude" workshop materials — nine hands-on courses on skills, MCP, evals and Claude Managed Agents.
- official
- workshops
- anthropic
- managed-agents
Anthropic's official starter kit for large-scale language migrations — the prompts, templates and scripts behind their documented process.
The companion kit to Anthropic's "How we run large-scale code migrations with Claude Code": defaults to structure-preserving migrations (same architecture, same data structures, new language) with six templated steps from prompts/00-feasibility.md onward. A portable "judge" (your test suite, or one built by 00b-judge-setup.md) is the exit condition — no judge, no migration. Provenance is real: prompts were generalized and reviewed against production ports including Bun's 1M+ line Zig→Rust move.
For teams staring down a cross-language port (Zig→Rust, JS→TS, Python→Go scale). If your architecture stays and only the language changes, this kit turns a scary program into a checklist; the judge-first framing alone prevents the classic "we translated 80% and now nothing works" failure.
1) Clone inside or adjacent to the repo being migrated. 2) Copy or @-import the kit's CLAUDE.md into your repo's CLAUDE.md. 3) Optional: cp -r migration-kit/skill ~/.claude/skills/code-migration (then fix [kit path] inside). 4) Open Claude Code in the target repo, fill placeholders in prompts/00-feasibility.md, paste.
Reference code, not actively maintained; issues and PRs not monitored. Designed for structure-preserving moves — if you're redesigning as you migrate, the README flags that parts of the kit change and one part becomes invalid.
Charted
Anthropic's official "Code with Claude" workshop materials — nine hands-on courses on skills, MCP, evals and Claude Managed Agents.
Anthropic's announcement of Fable 5.1 and Mythos 5.1 — one base model, two guardrail postures.
Anthropic's canonical essay on agent design — workflows vs agents, and five composable patterns.