review-a-skill

1# Review A Skill
2 
3## Goal
4 
5Evaluate a single skill's quality, relevance, and safety before it enters or stays in a library.
6 
7## Guardrails
8 
9- Always use `--format json` for machine-readable output in automated pipelines.
10- Always use `--fields` to limit output size when inspecting catalog entries.
11- Always use `--dry-run` before curating or removing a skill.
12- Never remove a skill without first checking if other skills depend on it via `info --format json` dependencies.
13 
14## Workflow
15 
161. Preview the skill content to check for quality and safety.
17 
18```bash
19npx ai-agent-skills preview <skill-name>
20```
21 
22The preview command sanitizes content — if it flags sanitization, investigate before proceeding.
23 
242. Inspect the catalog entry for metadata completeness.
25 
26```bash
27npx ai-agent-skills info <skill-name> --format json --fields name,description,tags,collections,dependencies
28```
29 
303. Validate the skill's SKILL.md structure.
31 
32```bash
33npx ai-agent-skills validate <skill-name>
34```
35 
364. If the skill needs curation (notes, collections, verification):
37 
38```bash
39npx ai-agent-skills curate <skill-name> --notes "Reviewed: solid patterns" --verify --dry-run
40npx ai-agent-skills curate <skill-name> --notes "Reviewed: solid patterns" --verify
41```
42 
435. If the skill should be removed:
44 
45```bash
46npx ai-agent-skills curate <skill-name> --remove --dry-run
47npx ai-agent-skills curate <skill-name> --remove --yes
48```
49 
50## Decision Criteria
51…