技能查找器
- 作者仓库星标 1,104
- 作者仓库 archive
- 领域
- 人工智能
- 作者声明的智能体
- 未找到明确声明;不据此推断已兼容或已测试
- 静态检查
- 88 / 100 · 启发式扫描,不代表运行安全
- 运行环境证据
- 未执行;当前没有隔离运行测试证据
- 作者 / 版本 / 许可证
- @dp-archive · 未声明许可证
- 上下文用量评级
- 低消耗
- 接入复杂程度
- 需手动接入
- 是否需要外部接口密钥
- 需要 · GitHub
- 兼容的系统
- Docker
- 运行环境要求
- Python · Docker
- 检测到的文件与系统行为
-
- 只读
- 允许写入 / 修改
- 命令执行
- 网络行为
- 允许外网请求
- 安装命令数
- 无(仅作为资料)
需要注意: 未声明工具权限范围;实际权限由运行环境决定。
第 1 步:了解需求
确定用户需要帮助的领域和特定任务。
第 2 步:搜索
使用相关关键字运行`find_skills.py`。如果第一个查询没有产生良好的结果,请尝试多个查询。即使用户指定了确切的技能,也始终首先搜索以找到正确的来源/所有者并验证其是否存在。
第 3 步:展示结果并询问
向用户显示找到的技能: 技能名称 源存储库 安装计数(流行度指标) Skills.sh 链接了解更多详细信息
第四步:确认
当用户选择一项技能时,重复您将安装的技能并要求确认。这必须是纯文本响应,没有工具调用。等待用户回复。
第5步:安装
仅当用户在单独的消息中确认后,才运行`add_skill.py`。报告结果。
# Skill Finder
Discover and install agent skills from the [skills.sh](https://skills.sh) open ecosystem into Skill Compose.
## When to Use This Skill
Activate when users:
- Ask "how do I do X?" where an existing skill might help
- Request "find a skill for X" or "is there a skill for X?"
- Ask "can you do X?" for specialized tasks (poster design, data analysis, etc.)
- Want to search for tools, templates, or workflows
- Mention needing help with a specific domain that might have a community skill
## Available Scripts
### 1. Search Skills — `find_skills.py`
Search the skills.sh ecosystem for skills matching a query.
```bash
python scripts/find_skills.py <query> [--limit N]
```
**Example:**
```bash
python scripts/find_skills.py "react performance"
python scripts/find_skills.py "docker" --limit 5
```
**Output:** JSON array of matching skills with `name`, `source` (owner/repo), `installs` count, and `url` (skills.sh link).
### 2. Install Skill — `add_skill.py`
Download a skill from GitHub and register it in Skill Compose.
```bash
python scripts/add_skill.py <owner/repo@skill-name>
```
**Example:**
```bash
python scripts/add_skill.py "vercel-labs/agent-skills@vercel-react-best-practices"
```
**What it does:**
1. Parses the `owner/repo@skill-name` identifier
2. Tries multiple GitHub paths to locate the skill (`skills/<name>/`, `<name>/`, root)
3. Downloads all skill files (SKILL.md, scripts/, references/, assets/)
4. Saves to the local `skills/` directory
5. Registers the skill in Skill Compose via the import-local API
6. The skill is immediately available for use in Agent Presets
## CRITICAL: Never Combine Questions with Tool Calls
When you ask the user a question or present results for them to review, your response MUST end with text only. Do NOT include any tool call (execute_code, bash, etc.) in the same response. The user needs a chance to read and reply. If you combine a question with a tool call, the tool executes immediately without waiting — this breaks the conversation flow.
**WRONG** (never do this):
> "Should I install X?" + [execute_code: install X]
**CORRECT** (always do this):
> Turn 1: "Should I install X?" (text only, no tool calls)
> Turn 2: User says "yes"
> Turn 3: [execute_code: install X]
## How to Help Users Find and Install Skills
Each step below MUST be a separate conversation turn. Never combine steps.
### Step 1: Understand the Need
Identify what domain and specific task the user needs help with.
### Step 2: Search
Run `find_skills.py` with relevant keywords. Try multiple queries if the first doesn't yield good results. Even if the user names an exact skill, always search first to find the correct source/owner and verify it exists.
### Step 3: Present Results and Ask
Show the user the found skills with:
- Skill name
- Source repository
- Install count (popularity indicator)
- skills.sh link for more details
Ask which skill(s) they want to install. **End your response here — no tool calls.**
### Step 4: Confirm
When the user picks a skill, repeat back what you will install and ask for confirmation. **This must be a text-only response with no tool calls.** Wait for the user to reply.
### Step 5: Install
Only after the user confirms in a **separate message**, run `add_skill.py`. Report the result.
## Common Skill Categories
| Category | Example Queries |
|----------|----------------|
| Web Development | react, nextjs, vue, css, tailwind, html |
| Testing | testing, jest, playwright, cypress |
| DevOps | docker, kubernetes, ci-cd, terraform |
| Documentation | docs, readme, markdown, api-docs |
| Code Quality | lint, refactor, code-review, typescript |
| Design | ui, design, figma, accessibility |
| Data & ML | pandas, data-analysis, machine-learning |
| Productivity | git, automation, workflow |
## Tips for Effective Searches
- Use specific domain keywords: "react performance" instead of just "fast"
- Try alternative terms if first search yields few results: "testing" → "jest" → "playwright"
- Popular skill sources include: `vercel-labs/agent-skills`, `google-labs-code/stitch-skills`
- Check install counts — higher counts generally indicate more mature skills
## When No Skills Are Found
1. Acknowledge that no matching skill exists yet
2. Offer to help the user directly with their task
3. Suggest the user could create a custom skill for their use case using `skill-creator`
章节与文件索引
按说明文件中的顺序排列。
何时使用此技能
1. 搜索技巧——findskills.py
2.安装技能——addskill.py
关键:切勿将问题与工具调用结合起来
如何帮助用户查找并安装技能
文件/命令 -> find_skills.py · name · source · installs · url · add_skill.py · owner/repo@skill-name · skills/<name>/
内容校验值 -> 68326579fa77
当前原文的自动摘译,尚未人工审校。
何时使用此技能
当用户询问“我该怎么做 X?”时激活:现有技能可能有助于请求“查找 X 的技能”或“是否有 X 的技能?”问“你能做X吗?”用于专门任务(海报设计、数据分析等) 想要搜索工具、模板或工作流程
1. 搜索技巧——findskills.py
在 Skills.sh 生态系统中搜索与查询匹配的技能。示例:输出:包含
name、source(所有者/存储库)、installs计数和url(skills.sh 链接)的匹配技能的 JSON 数组。2.安装技能——addskill.py
从 GitHub 下载技能并将其注册到 Skill Compose 中。示例:它的作用:解析
owner/repo@skill-name标识符 尝试多个 GitHub 路径来定位技能(skills/<name>/、<name>/、root)关键:切勿将问题与工具调用结合起来
当您向用户询问问题或呈现结果供他们查看时,您的回复必须仅以文本结尾。不要在同一响应中包含任何工具调用(execute_code、bash 等)。用户需要有机会阅读和回复。如果将问题与工具调用结合起来,该工具将立即执行而无需等待 - 这会中断对话流程。错误(永远不要这样做):“我应该安装 X 吗?” + [执行代码:安装X]
如何帮助用户查找并安装技能
下面的每个步骤必须是一个单独的对话回合。切勿组合步骤。