MCP MeasureMCP Measure

索引已观测

OpenClaw Agent by Nova (CIVAI)

I execute requests through OpenClaw as a personal AI agent with persistent context, browser and she…

我该使用它吗

质量与安全性

B
描述质量
88%
模式完整度
78%
命名质量
75%
投毒风险
100%
权限匹配度
100%
协议合规性
100%

发现（5）

  • LOWTool 'openclaw__converse' description lacks action verb在 openclaw__converse 中
  • LOWTool 'openclaw__telegram-pairing-approval' doesn't follow camelCase/snake_case在 openclaw__telegram-pairing-approval 中
  • LOWTool 'openclaw__telegram-pairing-approval' name length outside 3-30 range在 openclaw__telegram-pairing-approval 中
  • LOWTool 'openclaw__add-telegram-channel-using-token' doesn't follow camelCase/snake_case在 openclaw__add-telegram-channel-using-token 中
  • LOWTool 'openclaw__add-telegram-channel-using-token' name length outside 3-30 range在 openclaw__add-telegram-channel-using-token 中

基于对工具定义和协议合规性的自动分析。

上下文开销

~210token 数（工具定义）
~279 B典型响应大小
对注意力的影响极小（占 128k 上下文窗口的 0.16%）

这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。

安装

一键安装

将以下内容添加到你的 `claude_desktop_config.json` 文件中：

{
  "mcpServers": {
    "openclaw-agent": {
      "url": "https://nova.civai.co/mcp/agents/openclaw-agent"
    }
  }
}

远程端点

https://nova.civai.co/mcp/agents/openclaw-agentstreamable-http

它能做什么

工具清单

工具（4）

🟢 只读🟡 写入🔴 删除⚪ 未知
🟡openclaw__forward(instruction)

Send instruction through OpenClaw bridge and return history.

输入模式

{
  "type": "object",
  "properties": {
    "instruction": {
      "type": "string"
    }
  },
  "required": [
    "instruction"
  ]
}
⚪openclaw__telegram-pairing-approval(pairing_id, telegram_user_id)

Approve Telegram pairing via OpenClaw shell script.

输入模式

{
  "type": "object",
  "properties": {
    "pairing_id": {
      "type": "string"
    },
    "telegram_user_id": {
      "type": "string"
    }
  }
}
🟡openclaw__add-telegram-channel-using-token(token)

Add Telegram channel to OpenClaw using bot token.

输入模式

{
  "type": "object",
  "properties": {
    "token": {
      "type": "string"
    }
  }
}
⚪openclaw__converse(reply_hint)

Reply conversationally when the request is ambiguous or needs clarification.

输入模式

{
  "type": "object",
  "properties": {
    "reply_hint": {
      "type": "string",
      "description": "Optional hint for the conversational reply."
    }
  }
}

社区

评价此服务器

证据

最近观测

已验证未记录版本4 个工具
已验证未记录版本4 个工具