CommandVault

Search prompts, CLAUDE.md/AGENTS.md templates, Cursor rules, Agent Skills and MCP servers.

使うべきか

品質と安全性

A
説明の品質
86%
スキーマの完全性
75%
命名の品質
96%
ポイズニングのリスク
100%
権限の一致
100%
プロトコルへの準拠
100%

検出事項(1)

  • LOWTool 'list_communities' description lacks action verblist_communities 内

ツール定義とプロトコルへの準拠に関する自動分析に基づいています。

コンテキストコスト

~629トークン数(ツール定義)
~790 B一般的なレスポンスサイズ
注意への影響は最小限(128k コンテキストの 0.49%)

これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。

インストール

ワンクリックインストール

これを `claude_desktop_config.json` ファイルに追加してください:

{
  "mcpServers": {
    "commandvault": {
      "url": "https://commandvault.ai/api/mcp"
    }
  }
}

リモートエンドポイント

https://commandvault.ai/api/mcpstreamable-http

できること

ツール一覧

ツール(5)

🟢 読み取り専用🟡 書き込み🔴 削除⚪ 不明
🟢search_commands(query, community, tag, kind, sort, ...)

Search CommandVault's library of AI commands (prompts, CLAUDE.md/AGENTS.md templates, Cursor rules, slash commands, subagents). Returns slugs to pass to get_command.

入力スキーマ

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Words to look for in titles, summaries and text. Leave empty to browse."
    },
    "community": {
      "type": "string",
      "description": "Community slug, e.g. 'agents', 'frontend', 'marketing'. See list_communities."
    },
    "tag": {
      "type": "string",
      "description": "Tag, e.g. 'claude-code', 'cursor', 'nextjs'."
    },
    "kind": {
      "type": "string",
      "enum": [
        "prompt",
        "workflow",
        "template",
        "snippet",
        "checklist",
        "skill",
        "mcp_server"
      ],
      "description": "Only this kind. 'skill' = SKILL.md Agent Skills, 'mcp_server' = MCP servers, 'template' = files like CLAUDE.md or Cursor rules."
    },
    "sort": {
      "type": "string",
      "enum": [
        "top",
        "hot",
        "new"
      ],
      "default": "top"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 25,
      "default": 10
    }
  }
}
🟢get_command(slug, variables)

Fetch the full text of a command by slug, with {{variables}} filled from the values you pass. Unfilled variables are listed so you can ask the user for them.

入力スキーマ

{
  "type": "object",
  "properties": {
    "slug": {
      "type": "string",
      "description": "The command's slug from search_commands or its URL (commandvault.ai/commands/<slug>)."
    },
    "variables": {
      "type": "object",
      "additionalProperties": {
        "type": "string"
      },
      "description": "Values for the command's {{blanks}}, keyed by variable name."
    }
  },
  "required": [
    "slug"
  ]
}
🟢list_communities

All CommandVault communities with their slug and number of commands.

入力スキーマ

{
  "type": "object",
  "properties": {}
}
🟢list_collections(slug)

Curated collections (starter kits such as 'claude-code-starter-kit' or 'cursor-rules-by-framework'). Pass a slug to get the commands in one.

入力スキーマ

{
  "type": "object",
  "properties": {
    "slug": {
      "type": "string",
      "description": "Optional collection slug to list its commands."
    }
  }
}
🟢trending_commands(limit)

The commands people are copying, running and voting for most this week.

入力スキーマ

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 25,
      "default": 10
    }
  }
}

コミュニティ

このサーバーを評価する

エビデンス

最近の観測

検証済みバージョンは記録されていませんツール 5 件