一键导入
setup
Set up and configure Skillet skill discovery. Use when the user wants to set up skillet, configure repos, or customize server behavior.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Set up and configure Skillet skill discovery. Use when the user wants to set up skillet, configure repos, or customize server behavior.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | setup |
| description | Set up and configure Skillet skill discovery. Use when the user wants to set up skillet, configure repos, or customize server behavior. |
| version | 2026.02.27 |
| trigger | Use when the user wants to set up skillet, configure skill discovery, or manage skill installation preferences |
| license | MIT OR Apache-2.0 |
| author | Josh Rotenberg |
| categories | ["tools","configuration"] |
| tags | ["skillet","skills","setup","mcp"] |
Skillet is an MCP-native skill discovery tool. It gives you access to a searchable library of agent skills at runtime, served as MCP prompts.
Add this to your .mcp.json (project-level) or ~/.claude/settings.json
(global):
{
"mcpServers": {
"skillet": {
"command": "skillet",
"args": ["serve"]
}
}
}
Or with Docker (no install needed):
{
"mcpServers": {
"skillet": {
"command": "docker",
"args": ["run", "-i", "--rm", "ghcr.io/joshrotenberg/skillet:latest"]
}
}
}
After adding, restart your agent to connect.
Once connected, skills are available as MCP prompts. Search for a skill, then use it as a prompt for the current session. No files written, no restart needed.
1. search_skills("rust development")
2. Use the skill prompt for joshrotenberg/rust-dev
3. Follow the skill's instructions
search_skills(query) -- search by keyword, with optional category,
tag, or model filterslist_categories() -- browse available categorieslist_skills_by_owner(owner) -- see all skills by a publisherWhen the user's task could benefit from a skill you don't have locally, proactively search skillet. For example, if asked to write Python code and you don't have Python conventions loaded, search for Python skills.
If search returns no results for the user's topic, check the
skillet/skill-repos skill for external repositories that may cover it.
Use info_skill("skillet/skill-repos") to see the curated list.
Skillet is configured via ~/.config/skillet/config.toml and CLI flags.
Adding custom repos:
skillet serve --remote https://github.com/org/skills.git
Skipping the official repo:
skillet serve --no-official-repo
Config file (~/.config/skillet/config.toml):
[repos]
remote = ["https://github.com/org/skills.git"]
[server]
discover_local = false
If the user has a preference for how skills should be used, respect it:
Curated directory of external skill repositories. Use when the user needs skills not available locally, or to suggest adding new skill sources.
Contributing to skillet itself. Covers architecture, development setup, testing, and PR workflow.
Creating and maintaining skill repos. Covers setup, directory layout, PR review, multi-repo, and hosting.
Authoring skills for skillet. Covers the skill format, discovery via suggest, and project manifests.
Using skillet as a skill consumer. Covers searching, browsing, and using skills as MCP prompts.