skill-visionlisted

# Skill Vision *Boards your Agent Skills and inspects every plank before they sail — validation, script testing, and quality scoring.* **Category**: Engineering Quality Assurance · **Dependencies**: None (Python stdlib only) Meta-skill that validates, tests, and scores skills in this repository. Tier is computed, not asserted: run `quality_scorer.py` and read `tier_recommendation`. Six tools, run from the **repo root** with full paths: 1. **`scripts/spec_validator.py`** — the Agent Skills spec: does it load, upload, trigger? 2. **`scripts/claim_auditor.py`** — does the documentation tell the truth about the code? 3. **`scripts/skill_validator.py`** — structure + documentation compliance 4. **`scripts/script_tester.py`** — Python script syntax/imports/runtime/output testing 5. **`scripts/quality_scorer.py`** — multi-dimensional scoring with letter grade 6. **`scripts/security_scorer.py`** — security posture scoring (also available via `quality_scorer.py --include-security`) 7. **`scripts/skill_mapper.py`** — draws the skill's files and reference graph as a Mermaid flowchart ## Is it true, not just well-formed? Every other tool here asks whether a skill is *structured* correctly. `claim_auditor.py` asks whether it is *honest*. Those fail independently: a skill can pass every link check, score an A, and still tell you in its frontmatter that a script runs offline when that script ships your prompt to a third-party API. ```bash python3 scripts/claim_auditor.py path/to/skil