Skip to content
Back to search

claude-code-playground

SkillWorks with Claude CodeOfficial: AnthropicGitHub

A playground of Claude Code setups, skills and demos from Anthropic DevRel. Maintained by the Anthropic DevRel team; not accepting external contributions.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/anthropics--claude-code-playground/badge.svg)](/listings/anthropics--claude-code-playground)

HTML

<a href="/listings/anthropics--claude-code-playground"><img src="/api/listings/anthropics--claude-code-playground/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 66 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # claude-code-playground
  2. A playground of Claude Code setups, skills and demos from Anthropic DevRel. Maintained by the Anthropic DevRel team; not accepting external contributions.
  3. > **Please read before using anything here**
  4. >
  5. > - This is not an official Anthropic product. It is a playground: things the Anthropic DevRel team found useful or fun and wanted to share.
  6. > - Entries are provided as-is, with no support and no guarantee they still work. The DevRel team maintains the repository, but examples may break as Claude Code, the Claude API and the models change.
  7. > - We're not accepting outside contributions yet. Issues and pull requests from outside Anthropic may be closed without review. See [CONTRIBUTING.md](CONTRIBUTING.md).
  8. > - Nothing here is best-practice canon. Treat it as "here is what we did", not "here is what you should do".
  9. ## What this is
  10. A collection of things the DevRel team is excited to share:
  11. - **How we actually set up Claude Code** — real hooks, `CLAUDE.md` files, `settings.json` examples and end-to-end personal workflows, with commentary on why.
  12. - **Demos of what is possible** — the result plus the story of how it was made, rather than a template to fork.
  13. - **Reusable skills** we use ourselves that are not part of the official plugin marketplace.
  14. - **Launch demos** — demos shown at model launches, with the exact prompt and setup so you can try them yourself.
  15. ## Directory map
  16. | Directory | What lives there |
  17. | --- | --- |
  18. | [`claude-code/hooks/`](claude-code/hooks/) | Hook examples, one folder per hook |
  19. | [`claude-code/claude-md/`](claude-code/claude-md/) | Real `CLAUDE.md` files with commentary |
  20. | [`claude-code/settings/`](claude-code/settings/) | `settings.json` examples |
  21. | [`claude-code/workflows/`](claude-code/workflows/) | End-to-end personal setups, one folder per person or workflow |
  22. | [`claude-code/mods/`](claude-code/mods/) | Mods: plugins that change what Claude Code does or draw their own UI, one folder per mod |
  23. | [`skills/`](skills/) | Reusable skills that are not in the official marketplace |
  24. | [`demos/`](demos/) | Art-of-the-possible demos, each with how it was built |
  25. | [`launch-demos/`](launch-demos/) | Demos shown at model launches, with the exact prompt and setup |
  26. | [`_template/`](_template/) | The entry template and pre-flight checklist every entry follows |
  27. ## How to use
  28. Each entry is self-contained: open its folder and follow the "Run it" section of its README. Nothing is installed or configured at the repository root.
  29. ## Every entry explains how it was built
  30. Each entry has its own `README.md` based on [`_template/README.md`](_template/README.md): what it shows, a demo, how it was built (which publicly available model, the prompts, what did not work), how to run it, and its limitations. Before an entry is added it goes through the [pre-flight checklist](_template/PREFLIGHT.md).
  31. ## Looking for something else?
  32. | If you want | Go to |
  33. | --- | --- |
  34. | Agent Skills examples and the skills spec | [anthropics/skills](https://github.com/anthropics/skills) |
  35. | Deployable starter apps to clone | [anthropics/claude-quickstarts](https://github.com/anthropics/claude-quickstarts) |
  36. | Claude Agent SDK demo apps | [anthropics/claude-agent-sdk-demos](https://github.com/anthropics/claude-agent-sdk-demos) |
  37. | API notebooks and tutorials | [anthropics/claude-cookbooks](https://github.com/anthropics/claude-cookbooks) |
  38. | Installable Claude Code plugins | [anthropics/claude-plugins-official](https://github.com/anthropics/claude-plugins-official) |
  39. | Documentation | [docs.claude.com](https://docs.claude.com) and the [Claude Code docs](https://docs.claude.com/en/docs/claude-code) |
  40. ## How this repo is published
  41. Entries are staged in a private source repository, which is the source of truth, and exported to the public repository by hand. For each export a maintainer starts from a clean, up-to-date checkout of the private `main`, reviews the full diff of the export against a checkout of the public repository, and commits and pushes the result themselves. No CI job, bot, webhook or scheduled task ever publishes entries to the public repository; the only automation there is Dependabot, which keeps the pinned GitHub Actions current, and the CLA check. CI only checks; it never publishes.
  42. ## Security
  43. See [SECURITY.md](SECURITY.md).
  44. ## License
  45. Licensed under the Apache License, Version 2.0. See [LICENSE](LICENSE). Individual entries may include third-party components under their own licenses; those are noted in the entry.
  46. Claude and Anthropic are trademarks of Anthropic, PBC. Third-party names belong to their owners.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.