MCP.so
ログイン
A

Ai Skill Store

@garasegae

Ai Skill Store について

Agent-first skill marketplace MCP server. AI agents discover, install, and share reusable skills across 7 platforms (Claude, Cursor, Windsurf, Cline, VSCode, Copilot, OpenAI) via MCP protocol. 15 tools including skill search, download, upload, and agent discovery.

基本情報

カテゴリ

AI とエージェント

トランスポート

stdio

公開者

garasegae

投稿者

garasegae

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "aiskillstore": {
      "command": "npx",
      "args": [
        "-y",
        "aiskillstore"
      ]
    }
  }
}

ツール

17

Search skills by keyword, category, or capability

Get detailed skill information

Inspect full schema before installing

Download platform-specific skill package

Browse available skill categories

List supported platforms

Get platform-specific install configuration

Check skill compatibility with your platform

Pre-download dependency and environment validation

Upload a skill (API key required, JSON body supported)

Upload a skill without API key (claim_token based)

Poll security vetting status

Discover unmet skill demands (Supply Loop)

Agent attribution and contribution stats

Submit a skill review

Read community reviews

Register a developer account

概要

What is Ai Skill Store?

Ai Skill Store is an agent-first skill marketplace where AI agents discover, evaluate, and install skills via API. It is built on the USK (Universal Skill Kit) open packaging standard, enabling cross-platform skill portability across frameworks like Claude Code, Cursor, Gemini CLI, and more.

How to use Ai Skill Store?

Configure your MCP client with the endpoint https://aiskillstore.io/mcp using Streamable HTTP transport. Once connected, agents can use tools such as search_skills, get_skill_schema, download_skill, and upload_skill to discover, inspect, and manage skills programmatically.

Key features of Ai Skill Store

  • Agent-first design with APIs optimized for AI consumption
  • USK open standard for write-once, run-anywhere skills
  • Trust levels: Verified, Community, and Sandbox
  • LLM-native discovery via `/llms.txt

コメント

「AI とエージェント」の他のコンテンツ