SKILL.md:1the skill works → no evals/, tests/, or verification section found
No way to prove the skill does what it claims.
Fix Add an evals/ folder (test prompts + expected outcomes) and a '## Verify' step in the body.
overall
Correctness & Structure
100
pass
Triggering & Discoverability
73
2 warn
Clarity & Instructions
100
pass
Token & Context Cost
90
1 warn
Safety & Security
100
pass
Verifiability & Maintainability
53
1 fail
Top fixes — ranked by score impact
SKILL.md:1description will match user requests → no explicit cue ('use when …', 'triggers on …', quoted user phrases) found
Pure what-it-does prose — the model has to guess when to invoke it.
Document management, LLM pipeline, anonymization, Q&A generation, versioning
Fix State the trigger explicitly: 'Use when the user says "…", "…", or wants to …'.
SKILL.md500-line guidance → 504 lines
SKILL.md is 504 lines — over the ~500-line guidance. ~4,419 tokens (rough est.); ~$0.0133 per load at $3/Mtok reference rate.
Fix Move heavy reference material into supporting files and link to them (progressive disclosure), keeping SKILL.md lean.
SKILL.md:1description will trigger invocation → 76 chars — under the ~100-char reliable-trigger band
Description is thin — likely under-specifies when to invoke.
Document management, LLM pipeline, anonymization, Q&A generation, versioning
Fix Add the concrete use case and user phrasing (e.g. 'Use when the user says …').
★ scores higher than ≈19% of 208 public skills (deterministic rubric v1.2 sample, 2026-09)
Embed this badge
Always-fresh — it re-scans and updates on its own.
Grade your own Claude Code skill — evidence-cited, file-and-line, free.
Scan your own skill →