Official description๐ ๐ ๐ ๐ช ๐ง - Search, install and manage agent skills (SKILL.md files) from GitHub repositories, with workspace analysis for personalized recommendations.
Config template (adjust for your client)
claude mcp add skill-ninja-mcp-server -- npx -y skill-ninja-mcp-server
A pasteable skeleton generated from the repository name. Check the project README for the exact package and arguments.
Star count
12 โ ยท aktsmm/skill-ninja-mcp-server ยท read from the repository page ยท (checked 2026-10-16 20:00:20)
Source
collected from the punkpeye/awesome-mcp-servers list ยท type: MCP server
Harvest history
first seen 2026-10-09 ยท last seen 2026-10-16 ยท seen 32ร ยท listed by 7 sources
Score
score 703 ยท 12 โ ยท 7 sources ยท mcp
FAQ (answers come from our records)
How do I install it?
We did not find an install command for this repository (our install coverage is only about 4%). Check its README or releases page.
Who is it for?
We filed it under these tags: MCP ๆๅก. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository aktsmm/skill-ninja-mcp-server๏ผfirst catalogued 2026-10-09๏ผlast seen 2026-10-16ใThere is no official certification or quality rating here โ we record sources and dates only.
ๆฅๆฌ่ช็ README An MCP (Model Context Protocol) server for searching, installing, and managing AI Agent Skills. Works with MCP-compatible clients like Claude Desktop, Cursor, and VS Code. Installation Or run it directly with npx: Configuration Claude Desktop ~/.claude/claude_desktop_config.json: VS Code (mcp.json) %APPDATA%\Code\User\mcp.json: Environment Variables | Variable | Description | Default | | -------------------------------- | --------------------------------------------------------- | ------------------------------------- | | GITHUB_TOKEN | GitHub API token for higher rate limits | none | | SKILL_NINJA_INDEX_DIR | Skill index storage directory | ~/.skill-ninja | | SKILL_NINJA_TRUSTED_WORKSPACES | Trusted workspace roots allowed for read/write operations | auto-detect current project root only | | LANG | Output language, for example ja_JP | system default | Security Workspace-mutating tools only operate inside trusted workspace roots. - By default, the server trusts the current working directory only when it looks like a project root. - To allow other locations, set the SKILL_NINJA_TRUSTED_WORKSPACES environment variable to one or more trusted roots separated by your OS path delimiter. - Requests outside trusted roots are rejected before any read, write, or delete occurs. - GitHub API and raw content fetches use a bounded timeout so network failures return control instead of hanging the MCP server indefinitely. Duplicate Skill Names - Search and recommendation results include the source name for each skill. - If multiple sources publish the same skill name, pass the optional source field to skillNinja_install or skillNinja_localize. - Installed skill listings include the recorded source, and the server refuses to overwrite an installed skill with the same name from a different source. - If a partial skill name matches multiple different skills, the install, localize, and uninstall flows now stop and ask for the exact skill name instead of picking the first match. Tools | Tool | Description | | ------------------------ | ---------------------------------------------------------- | | skillNinja_search | Search the local skill index by keyword | | skillNinja_install | Install a skill into a trusted workspace | | skillNinja_uninstall | Remove an installed skill from a trusted workspace | | skillNinja_list | List installed skills in a trusted workspace | | skillNinja_recommend | Recommend skills based on workspace contents |