skilld
All skills
tradermonty avatar

/dual-axis-skill-reviewer

@4d045f3
by tradermontytradermonty/claude-trading-skills2.9k stars
663

Review skills in any project using a dual-axis method: (1) deterministic code-based checks (structure, scripts, tests, execution safety) and (2) LLM deep review findings. Use when you need reproducible quality scoring for `skills/*/SKILL.md`, want to gate merges with a score threshold (for example 90+), or need concrete improvement items for low-scoring skills. Works across projects via --project-root.

Ask your Agent

Use this Skill: https://skilld.dev/gh/tradermonty/claude-trading-skills/dual-axis-skill-reviewer

This session only. Nothing lands on disk.

referencesscoring_rubric.md

≈443 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Scoring Rubric Rationale

This document explains why each auto-axis component has its current weight.

Weights

  • metadata_use_case (20) Why: bad frontmatter and unclear trigger conditions make a skill hard to invoke correctly.

  • workflow_coverage (25) Why: operators need executable flow; missing core sections creates ambiguity in real use.

  • execution_safety_reproducibility (25) Why: command examples and path hygiene determine whether results are repeatable and safe.

  • supporting_artifacts (10) Why: scripts/references/tests provide baseline maintainability, but should not dominate scoring.

  • test_health (20) Why: runtime confidence is critical; passing tests strongly increases trust in automation.

Threshold Policy

  • 90+: production-ready baseline
  • 80-89: usable with targeted improvements
  • 70-79: notable gaps; strengthen before regular use
  • <70: high risk; treat as draft and prioritize fixes

Improvement Trigger

When final score is < 90, improvement items are mandatory in report output.

Declared Verification Is Not Scored

If the target project declares verification axes in skills-index.yaml, the reviewer surfaces them separately. They do not change component weights, findings, improvement items, or the final score. The declarations are metadata supplied by the target project, not an independent reviewer finding.

Knowledge-Only Skill Handling

For skills with no executable scripts (scripts/*.py absent) but with reference docs:

  • Classify as knowledge_only.
  • Do not penalize missing bash command examples.
  • Treat script/test artifacts as mostly not-applicable (supporting_artifacts and test_health adjusted).
  • Still require clear When to Use, Prerequisites, and workflow structure.

Source: SKILL.md on GitHub

Third-party checks

1 warning5d5 checks · Risk SAFE
  • Gen Agent Trust Hub5d

    The skill is a utility for auditing AI agent skills, performing automated quality checks and facilitating deep LLM reviews. It safely handles external input using safe YAML parsing and includes security-positive features like PII detection. While it can execute tests on evaluated projects, this is a core documented function performed through secure subprocess calls.

  • Socket5d

    No alerts

  • Snyk5d

    Risk: LOW · No issues

  • Runlayer7mo

    2/6 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Provenance

Signed by skilld at 4d045f3. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 17 hours ago.

Activeupdated 2 months ago

README badge

README badge for tradermonty/claude-trading-skills/dual-axis-skill-reviewer