agentleFS
Sign inStart free
komlukkomluk/scaffolding/skills/quality-validation/SKILL.md

quality-validation

Skill16 starsChanged 5 days agoMITView on GitHub

Score ResearchPacks (>=80) and Implementation Plans (>=85) against quality gates. TRIGGER when: gating research or a plan before implementation. SKIP: writing or running tests (use testing-strategy); security threat review (use security-review-checklists); verifying finished work by executing it (use verification-before-completion).

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All URLs valid, official sources: 10 pts

    line 46
  2. Distinctive:

    Every API has source URL: 10 pts

    line 49
  3. Distinctive:

    All file changes listed: 10 pts

    line 60
  4. Distinctive:

    Test plan included: 5 pts

    line 63
  5. Very specific:

    Run your project's own validate/lint/type-check/test command at the completion gate.

    line 81

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Quality Validation Skill

Systematic validation ensuring ResearchPacks and Implementation Plans meet quality standards before implementation.

Auto-Invoke Triggers

  • ResearchPack completion requires validation
  • Implementation Plan requires validation before coding
  • User requests quality check
  • Proceeding to next workflow phase (quality gate)

Quality Gates

Gate 1: Research → Planning

  • Trigger: ResearchPack completion
  • Pass: Score ≥ 80
  • Fail: Block, return defects

Gate 2: Planning → Implementation

  • Trigger: Implementation Plan completion
  • Pass: Score ≥ 85 AND APIs match
  • Fail: Block, return defects
  • Note: the ≥ 85 plan gate / design review may run on the opus tier (architect/analyst on model: opus, effort: high) for higher rigor. See docs/model-tiers.md.

Gate 3: Implementation → Completion

  • Trigger: Code complete
  • Pass: your project's validate/lint/test command passes (e.g. npm run validate, pytest, make check, cargo test)
  • Fail: Trigger self-correction (max 3 attempts)

ResearchPack Validation (100 pts, pass: 80)

Completeness (40 pts)

  • Library/API with version: 10 pts
  • 3+ key APIs documented: 10 pts
  • Setup/configuration steps: 10 pts
  • 1+ complete code examples: 10 pts

Accuracy (30 pts)

  • API signatures match official docs: 15 pts
  • Version numbers correct: 5 pts
  • All URLs valid, official sources: 10 pts

Citation (20 pts)

  • Every API has source URL: 10 pts
  • Version/section references: 5 pts
  • Confidence level stated: 5 pts

Actionability (10 pts)

  • Implementation checklist: 5 pts
  • Open questions identified: 5 pts

Implementation Plan Validation (100 pts, pass: 85)

Completeness (35 pts)

  • All file changes listed: 10 pts
  • Step-by-step sequence: 10 pts
  • Verification per step: 10 pts
  • Test plan included: 5 pts

Safety (30 pts)

  • Rollback plan complete: 15 pts
  • Risk assessment (3+ risks): 10 pts
  • Minimal changes: 5 pts

Clarity (20 pts)

  • Actionable steps: 10 pts
  • Success criteria defined: 5 pts
  • Time estimates: 5 pts

Alignment (15 pts)

  • Plan matches ResearchPack: 10 pts
  • Addresses all requirements: 5 pts

Validation Commands

Run your project's own validate/lint/type-check/test command at the completion gate. The exact command depends on your stack — substitute the equivalents below.

StepGeneric intentExample (Node/TS)Example (Python)
Type checkStatic type safetynpm run type-checkmypy .
LintStyle / static analysisnpm run lintruff check .
BuildCompile / packagenpm run buildpython -m build
TestsBehavioral verificationnpm testpytest
Full gateOne command runs allnpm run validatemake check
# Example (Node/TypeScript): full validation
npm run validate

# Example (Python): full validation
pytest && ruff check . && mypy .

Common Failures

FailureDescription
Hallucinated APIsAPIs not in official docs
Version mismatchWrong version in research
Missing citationsNo source URLs
No rollback planMissing reversion procedure
Ambiguous stepsNon-actionable instructions
API misalignmentPlan uses undocumented APIs

Validation Output Format

## Validation Report

**Artifact**: [ResearchPack|ImplementationPlan]
**Score**: XX/100
**Result**: [PASS|FAIL]

### Scoring Breakdown
- Completeness: XX/40
- Accuracy: XX/30
- Citation: XX/20
- Actionability: XX/10

### Defects Found
1. [Defect description]
2. [Defect description]

### Recommendations
- [Fix suggestion]

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free