# Skill Discovery

> 技能发现工具。搜索本地和云端技能库，发现可复用技能。当用户说"找技能"、"搜索技能"、"有什么技能"时使用。

- Skill: `wike-chi/skill-discovery` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add wike-chi/skill-discovery`
- Raw SKILL.md: https://api.skillmd.com/api/skills/wike-chi/skill-discovery/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: wike-chi (https://skillmd.com/u/wike-chi)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/wike-chi/skill-discovery

---


# Skill Discovery

> **Skill Graph：** 领域 → [[_index-meta|系统元技能领域]]


Discover and browse skills from OpenSpace's local and cloud skill library.

## When to use

- User asks "what skills are available?" or "is there a skill for X?"
- You encounter an unfamiliar task — a proven skill can save significant tokens over trial-and-error
- You need to decide: handle a task yourself, or delegate to OpenSpace

## search_skills

```
search_skills(query="automated deployment with rollback", source="all")
```

| Parameter | Required | Default | Description |
|-----------|----------|---------|-------------|
| `query` | yes | — | Natural language or keywords |
| `source` | no | `"all"` | Local + cloud; falls back to local-only if no API key |
| `limit` | no | `20` | Max results |
| `auto_import` | no | `true` | Auto-download top cloud hits locally |

## After search

Results are returned to you (not executed). Cloud hits with `auto_imported: true` include a `local_path`.

```
Found a matching skill?
├── YES, and I can follow it myself
│     → read SKILL.md at local_path, follow the instructions
├── YES, but I lack the capability
│     → delegate via execute_task (see delegate-task skill)
└── NO match
      → handle it yourself, or delegate via execute_task
```

## Notes

- This is for **discovery** — you see results and decide. For direct execution, use `execute_task` from the `delegate-task` skill.
- Cloud skills have been evolved through real use — more reliable than skills written from scratch.
- Always tell the user what you found (or didn't find) and what you recommend.