bdsqqq/dots/building-review-playgrounds
Claude Code and agent skill by bdsqqq/dots
- community
Checking that the SKILL.md file is still in the repository.
Exposes hard-to-reach UI states through issue-scoped production-component fixtures. Use when existing routes or stories cannot reproduce states needed for review.
Install
npx vskill@latest install bdsqqq/dots/building-review-playgroundsSKILL.md
building review playgrounds
build the smallest surface that reproduces the issue, not a second implementation.
contract
- input: issue or bounded change, production components/consumers, required states, available local data, and removal condition.
- output: reproducible route/story/fixture, state selectors or instructions, explicit fixture boundaries, and a list of temporary files or gates.
- termination: required states render through production code in representative layout, limitations are explicit, and removal is either verified or assigned to an explicit acceptance condition. capture/review claims belong to
producing-visual-evidence.
procedure
- try an existing application route or component story first. add a playground only when missing data, permissions, or setup prevents exercising required states. do not create one for a single readily reachable state.
- place it within the repository's preview conventions and scope it to the issue. prefer a local in-layout fixture when surrounding table, sidebar, scroll container, or flex context determines behavior. a helper demo does not establish that its real consumer works.
- import production components and styles. fixture only the unavailable inputs or external boundary; preserve the rendering and interaction under review. list what is synthetic and what therefore remains unverified. use fake data, never copied customer secrets.
- select states from the changed contract: ordinary content plus the long, empty, disabled, error, permission, or narrow-layout cases that could fail differently. preserve semantic distinctions instead of making every fixture the happy path. provide stable controls/selectors so the same state is repeatable.
- keep fixtures out of ordinary user navigation and production data paths. a query parameter alone is not an access-control boundary. prefer local-only wiring; if a committed route is necessary, use the application's supported development guard and verify production exclusion. never weaken authentication to make capture easier.
- record how to start and reach the fixture, its state-to-consumer mapping, and removal criteria. after review, remove temporary wiring and data within authorized scope; check imports, routes, build inputs, and the diff for residue. move genuinely useful regression coverage into the existing test/story system rather than retaining a migration museum.
evaluation
given a table cell that clips only with long content, a standalone centered component is insufficient. load the real cell in its constrained container, inject long fixture data, and confirm the expected wrapping/overflow state is reachable. inspect the rendered result, not just route compilation. reload and reproduce it without manual database edits.
verify the ordinary route still uses ordinary data, and that a production build cannot expose local-only fixtures. after removal, run the relevant build/type check and search for the fixture identifier/imports. report any retained fixture and its owner/removal condition; never claim cleanup from an empty screenshot.
provenance
depot migration: missing workflow data needed temporary in-layout fixtures, later removed. normative: meaningful contribution, skill design. constraints above stand without opening these links.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingmodules/agents/skills/building-review-playgrounds/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/bdsqqq/dots/building-review-playgrounds)