본문으로 건너뛰기
cccskills
무료GitHub에 공개

eval-harness

Evaluate agent and skill outputs against expected results. Score accuracy, completeness, and consistency, compare performance with a baseline, and use regression cases to spot quality drops after changes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

원문원문 설명 보기

Evaluation harness for testing agent and skill quality through structured benchmarks, regression tests, and quality scoring.

a5c-ai/babysitter

실행 점수 미검증

설치 방법 보기

이럴 때 유용합니다

  • Checking a new agent or skill against expected outputs
  • Finding regressions after changing instructions
  • Running periodic quality audits
  • Reviewing a skill before production use

포함된 파일(2)

  • SKILL.md2.1 KB
  • README.md230 B

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

  • Define test cases with known-correct outputs
  • Run agent against each test case
  • Score: accuracy, completeness, relevance
  • Compare against baseline performance
  • Track performance over time

2. Skill Quality Testing

  • Verify skill instructions produce expected outcomes
  • Test edge cases and boundary conditions
  • Measure consistency across multiple runs
  • Check for harmful or incorrect outputs
  • Validate against ground truth

3. Regression Suite

  • Collection of previously-passing test cases
  • Run after any agent/skill modification
  • Flag regressions with before/after comparison
  • Maintain pass rate threshold (>= 95%)

4. Process Verification

  • End-to-end process execution with known inputs
  • Verify each phase produces expected outputs
  • Check task ordering and dependency satisfaction
  • Measure total execution time

Quality Scoring

Accuracy Score (0-100)

  • Correctness of output vs expected
  • Partial credit for partially correct outputs
  • Penalty for hallucinated or fabricated content

Completeness Score (0-100)

  • Coverage of required output elements
  • Missing sections flagged and scored
  • Bonus for useful additional context

Consistency Score (0-100)

  • Run same input 3 times
  • Compare outputs for semantic similarity
  • Flag inconsistencies

Composite Score

  • (accuracy * 0.4 + completeness * 0.3 + consistency * 0.3)
  • Threshold: 80 to pass

When to Use

  • After creating new agents or skills
  • After modifying existing agents or skills
  • Periodic quality audits
  • Before promoting skills to production

Agents Used

  • Used by process-level evaluation orchestrators
  • No specific agent dependency (evaluates other agents)

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

advanced-ds-library

무료

Get implementations and guidance for advanced data structures used in competitive programming and complex algorithm problems. The skill covers balanced trees, persistent structures, tree decompositions, and range-query tools, with usage notes and complexity information.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Implementing a treap or splay tree
  • Answering range queries with a wavelet tree
  • Handling dynamic tree operations
a5c-ai/babysitter1.9천2026년 9월 17일 업데이트

adversarial-review

무료

Use this skill for an independent, evidence-based code review focused on whether an implementation meets its specification. It returns a clear PASS or FAIL verdict with file and line references, using a fresh reviewer for each re-review.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Checking whether code meets a specification
  • Running a final cross-unit code review
  • Re-reviewing a change after it failed
a5c-ai/babysitter1.9천2026년 9월 17일 업데이트

agent-booster

무료

Apply supported repetitive code edits with WebAssembly (WASM), a portable runtime, when a code pattern matches. Transforms include modernizing variable declarations, adding types or error handling, and converting Promise chains; unmatched patterns go to a coder fallback.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Modernizing repeated variable declarations
  • Adding TypeScript annotations to matching functions
  • Wrapping simple calls in error handling
a5c-ai/babysitter1.9천2026년 9월 17일 업데이트

agent-coordination

무료

Coordinates long-lived Crew agents and transient Polecat workers by assigning work through hooks, handing tasks between agents, and nudging stalled sessions. It also explains a three-level hook hierarchy and provides commands for listing, feeding, and reviving agents.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Assign work to persistent and short-lived agents
  • Hand off tasks between agents and prompt stalled workers
  • Set up shared, role-specific, and rig-specific hooks
a5c-ai/babysitter1.9천2026년 9월 17일 업데이트

algorithm-visualizer

무료

Turn algorithm execution into visual steps, showing how arrays, trees, graphs, and dynamic programming tables change. Generate formats such as ASCII, SVG, GIF, or HTML to explain concepts, inspect behavior, or support debugging.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Explaining sorting or search step by step
  • Showing graph traversal or shortest-path execution
  • Visualizing dynamic programming table changes
a5c-ai/babysitter1.9천2026년 9월 17일 업데이트

anti-drift

무료

Coordinate multi-agent work with short task cycles and regular checkpoints. The skill checks whether agents share consistent state and stay within assigned roles, then scores drift and guides corrections or human escalation when needed.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Coordinating agents on related subtasks
  • Keeping high-risk work aligned with its scope
  • Checking shared memory consistency
a5c-ai/babysitter1.9천2026년 9월 17일 업데이트

a5c-ai의 스킬 모두 보기

이 스킬의 문제 신고