Use when evaluating whether a skill belongs in a library. Preview content, check frontmatter, validate structure, and decide whether to keep, curate, or remove.
SkillScore77Maintained
Automatic
◆AI Engineer pack
Last verified 8/16/2026
What it changes
Passed 15 of 19 test cases with the skill, and 5 of 19 without it.
Without skill27%
With skill77%
Checks it passed that the model alone missed
Lists preview as the first step in the workflow sequence.
Specifies passing the --notes flag to the curate subcommand.
Specifies passing the --verify flag to the curate subcommand.
and 9 more in the full report
11 fixed6 passed either way5 still failing0 broke
Pass-rate lift
+50 pp
Tokens
-25%
Agent turns
0%
Response time
-63%
Model gemini-3.6-flashMeasured 8/16/2026v117 of 22 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 10
Prompt
We verified that 'vulnerable-parser' has no dependent skills and must be removed. What command variant should be run immediately prior to executing the final destructive removal command?
What was checked, without and with the skill
Checked
Without skill
With skill
Recommends executing curate with --remove and --dry-run before final deletion.
Missed
Passed
case-14 · gemini-3.6-flash · 8/16/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md60 linessingle file
Review A Skill
Goal
Evaluate a single skill's quality, relevance, and safety before it enters or stays in a library.
Guardrails
Always use --format json for machine-readable output in automated pipelines.
Always use --fields to limit output size when inspecting catalog entries.
Always use --dry-run before curating or removing a skill.
Never remove a skill without first checking if other skills depend on it via info --format json dependencies.
Workflow
Preview the skill content to check for quality and safety.
npx ai-agent-skills preview <skill-name>
The preview command sanitizes content — if it flags sanitization, investigate before proceeding.
Inspect the catalog entry for metadata completeness.
npx ai-agent-skills info <skill-name> --format json --fields name,description,tags,collections,dependencies
Validate the skill's SKILL.md structure.
npx ai-agent-skills validate <skill-name>
If the skill needs curation (notes, collections, verification):