Validate AI skills with pass/fail checks across ANALYZE, SMOKE, MANUAL, and RECOMMEND phases.
npx skills add https://github.com/demithras/ClearMeta --skill skill-check
Please help me install this Agent Skill. Skill: skill-check Source: https://github.com/demithras/ClearMeta/tree/main/.claude/skills/skill-check Command: npx skills add https://github.com/demithras/ClearMeta --skill skill-check
This Skill provides a structured, transparent framework for evaluating and validating the quality of AI skills using a layered checklist approach, replacing opaque scoring with clear pass/fail criteria.
/retro handoffs and skill-forge lifecycle management.Run the unified flow on a skill: /skill-check <skill-name> to begin analysis and validation.
High-intent search queries and answers about installing and using this skill.
Validate skill quality by applying structured checklist phases—ANALYZE, SMOKE, MANUAL, and RECOMMEND—that replace vague scores with verifiable pass/fail criteria. This enforces frontmatter validation and CLEAR-based exit reporting for consistent results.
Epistemic labeling with L-labels distinguishes automated checks from human-validated checks within the validation workflow. This tagging ensures transparent tracking of which lifecycle quality tests require manual verification versus automated execution.
Run a tiered validation workflow by executing a unified flow that adapts checks to lifecycle status—Draft, Active, or Production. The process applies consistent validation across all phases, ensuring appropriate rigor for each stage.
Skill validation supports automated coverage through tests.json driven validation. This configuration file defines automated checks that feed into the SMOKE and ANALYZE phases, ensuring consistent, repeatable quality verification.
You can log validation history and hand off results using integration hooks for retro handoffs and skill-forge lifecycle management. These hooks capture validation outcomes and feed them into continuous improvement tracking.
The best way to analyze skill quality without opaque scoring is using a layered checklist approach with CLEAR-based exit reporting. This framework provides transparent, verifiable criteria across ANALYZE and MANUAL phases instead of arbitrary numerical scores.
erenisiklarCreate Semgrep rules with YAML and test files for validation.
ribatshepoParse errors into structured diagnostic reports with redacted secrets and history tracking.
pradyumnasagarAutomate inter-skill data transfer and provenance capture via JSON Material Passports.
MuhammadUsmanGMScores Claude skills across nine quality categories and generates a structured validation report with remediation guidance.