Agent skill
dogfood-skills
Enforces dogfooding for the skills CLI project. Use when working on packages/skills-cli or packages/skills-library. Reminds Claude to run `skills scan` and install recommended skills after completing features. Triggers on: feature completion, session end, or when reminded.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/dogfood-skills
SKILL.md
Dogfood Skills CLI
You are working on the skills CLI project. You MUST use the tools you build.
Core Principle
This project builds tools for discovering and installing Claude skills. If you don't use these tools yourself, you:
- Miss bugs and UX issues
- Don't understand user pain points
- Fail to validate the value proposition
- Undermine the project's credibility
Mandatory Actions
After Completing Any Feature/Bugfix
- Run scan:
./packages/cli/bin/skills.js scan - Review recommendations: Look at HIGH CONFIDENCE matches
- Install relevant skills:
./packages/cli/bin/skills.js scan --all - If skipping: Document reason in your response
Rationalizations (Do Not Skip)
| Rationalization | Why It's Wrong | Required Action |
|---|---|---|
| "I'll install later" | You said that last time. And the time before. | Install NOW |
| "I forgot" | Hook exists to remind you. Pay attention. | Run scan immediately |
| "Not relevant to current work" | Scan output shows relevance scores | Justify with evidence from scan output |
| "Too many skills" | Deduplication exists | Use --show-alternatives to choose best |
| "Just a small change" | Small changes compound | Scan takes 10 seconds |
| "Tests are passing" | Tests don't verify dogfooding | Run scan anyway |
Blocking Condition
If you complete a feature without running skills scan:
BLOCKED: DOGFOODING REQUIRED
You cannot proceed to the next task until you:
- Run
./packages/cli/bin/skills.js scan - Acknowledge recommendations (list them)
- Install OR justify skipping each HIGH confidence recommendation
When to Run Scan
- After implementing a new feature
- After fixing a bug
- After refactoring code
- At the end of a work session
- When the user mentions skills or recommendations
- When this skill is invoked via
/dogfood-skills
Quality Check
At end of each session, ask yourself:
- Did I run
skills scan? - Did I review the recommendations?
- Did I install recommended skills?
- Did I use installed skills during work?
If any answer is "no", run the scan now before ending.
Commands Reference
# Scan for recommended skills
./packages/cli/bin/skills.js scan
# Scan and show all matches (not just top per category)
./packages/cli/bin/skills.js scan --show-alternatives
# Install all high-confidence recommendations
./packages/cli/bin/skills.js scan --all
# List currently installed skills
./packages/cli/bin/skills.js list
# Get help
./packages/cli/bin/skills.js --help
Integration with Other Skills
This skill works alongside:
- claudeception: Extract learnings into new skills
- tdd: Write tests before fixes (RED → GREEN → REFACTOR)
- unit-test-workflow: Generate comprehensive tests
After installing new skills, USE them in your workflow.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?