.claude/skills/skill-discovery/SKILL.md
Search external skill sources and evaluate candidates for reuse before creating custom skills
npx skillsauth add chemistrywow31/A-Team Skill DiscoveryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Search external skill repositories and marketplaces to find existing skills that match capability requirements. Evaluate candidates using a weighted scoring system and classify them for reuse, reference, or discard. This skill enforces the "reuse before create" principle.
This skill is used by the following agents:
skill-planner: Search external sources during Phase 2 planning to identify reusable skills before designing custom onesRead skills/skill-discovery/config.json if it exists. If not found, use these defaults:
{
"sources": {
"skillsmp": { "enabled": true },
"aitmpl": { "enabled": true },
"github": {
"enabled": true,
"repos": ["anthropics/skills"]
}
},
"search": {
"max_results_per_source": 5,
"min_score_threshold": 3.5
}
}
For each capability requirement extracted from the role design document:
Execute searches against all enabled sources using WebSearch. No API keys are required.
site:skillsmp.com {keywords} skillskillsmp.com/skills/site:aitmpl.com {keywords} skillaitmpl.com/component/skill/github.com SKILL.md "claude code" {keywords}site:github.com anthropics/skills {keywords} (and any repos listed in config)github.com/{owner}/{repo}/blob/main/ to raw.githubusercontent.com/{owner}/{repo}/main/)If the same skill appears in multiple sources, keep the entry with the most complete information. Note all sources where it was found.
Score each candidate skill on a 1-5 scale across four dimensions:
| Dimension | Weight | 1 (Poor) | 3 (Adequate) | 5 (Excellent) | |-----------|--------|-----------|---------------|----------------| | Relevance | 40% | Tangentially related | Partially matches the requirement | Directly matches the capability need | | Quality | 25% | No structure, no examples | Has structure but missing examples | Complete SKILL.md with frontmatter, sections, and examples | | Freshness | 15% | Over 12 months since last update | Updated within 6-12 months | Updated within 6 months | | Adoption | 20% | No stars, no platform listing | Listed on one platform or 10+ stars | Listed on multiple platforms or 50+ stars |
Weighted score = (Relevance x 0.4) + (Quality x 0.25) + (Freshness x 0.15) + (Adoption x 0.2)
| Score Range | Decision | Integration Pattern | |-------------|----------|---------------------| | >= 3.5 | Recommend reuse | Pattern A (Direct Install) or Pattern B (Adapted Install) | | 2.5 - 3.4 | Reference material | Pattern C (Reference Only) — do not install, provide to Skill Writer as reference | | < 2.5 | Discard | Create custom skill from scratch |
Use when the external skill scores >= 3.5 and requires no modifications:
name and description)skills/{skill-name}/SKILL.mdUse when the external skill scores >= 3.5 but requires modifications:
skills/{skill-name}/SKILL.mdUse when the external skill scores 2.5-3.4:
Every installed external skill (Pattern A or B) must include this section at the end of the SKILL.md:
## Source Attribution
- **Origin**: {source platform} ({URL})
- **Integration**: {Pattern A: Direct Install | Pattern B: Adapted Install}
- **Retrieved**: {date}
- **Modifications**: {None | list of changes made}
The search results must be reported in this structure:
## External Skills Discovery
### Search Summary
- Sources searched: {list of enabled sources}
- Total candidates found: {number}
- Recommended for reuse: {number}
- Reference materials: {number}
- Discarded: {number}
### Recommended External Skills
#### {skill-name} (Score: {x.x})
- **Source**: {platform} — {URL}
- **Relevance**: {score}/5 — {brief justification}
- **Quality**: {score}/5 — {brief justification}
- **Freshness**: {score}/5 — {brief justification}
- **Adoption**: {score}/5 — {brief justification}
- **Integration**: Pattern {A|B}
- **Target agents**: {agent-1}, {agent-2}
### Reference Materials
#### {skill-name} (Score: {x.x})
- **Source**: {URL}
- **Useful content**: {what can be referenced}
- **Target capability**: {which capability requirement this relates to}
Capability requirement: "Structured content writing for blog posts"
### Recommended External Skills
#### structured-writing (Score: 3.8)
- **Source**: SkillsMP — https://skillsmp.com/skills/structured-writing
- **Relevance**: 4/5 — Directly addresses content structuring methodology
- **Quality**: 4/5 — Complete SKILL.md with frontmatter and examples
- **Freshness**: 3/5 — Last updated 8 months ago
- **Adoption**: 4/5 — Listed on SkillsMP, 35 GitHub stars
- **Integration**: Pattern A
- **Target agents**: content-writer, editor
### Reference Materials
#### blog-seo-basics (Score: 3.0)
- **Source**: https://github.com/example/blog-seo-skill
- **Useful content**: SEO checklist and keyword density guidelines
- **Target capability**: SEO optimization checking
data-ai
Evaluate whether a proposed agent graph supports safe handoffs and useful parallel work
testing
Extract precise team requirements through staged, single-direction questioning
content-media
Search external sources for reusable skills before designing a new one
data-ai
Create or improve a skill by using the system skill when available and the legacy local bundle as fallback