MCP.so
登录

Muntuai Mcp

@design-smith

关于 Muntuai Mcp

AI agents can now launch and manage personalized outbound campaigns

基本信息

分类

其他

运行时

python

传输方式

stdio

发布者

design-smith

提交者

Zeke

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "muntu": {
      "url": "https://api.muntuai.com/api/mcp",
      "headers": {
        "Authorization": "Bearer ${env:MUNTU_AGENT_KEY}"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Muntuai Mcp?

Muntuai Mcp is the official Model Context Protocol (MCP) server for MuntuAI, an AI-native email outreach platform. It gives AI agents programmatic access to email outreach campaigns, lead management, domain infrastructure, and real-time workspace events through a standard MCP interface over HTTP.

How to use Muntuai Mcp?

Obtain an agent key from app.muntuai.com (Settings → Agent Keys), then configure your MCP client (Claude Desktop, Cursor) with the server URL https://api.muntuai.com/api/mcp and the Authorization: Bearer header using the key. Optionally set MUNTU_SESSION_ID to group requests into a logical session for safety controls and budgeting.

Key features of Muntuai Mcp

  • Read‑only resources: workspace, domains, senders, campaigns, campaign plans, events.
  • 30 tools grouped by domain: campaigns, leads, senders, domains, email generation, webhooks.
  • Three autonomy levels: observer, copilot, autonomous – control what agents can do.
  • Streamable HTTP transport with JSON-RPC 2.0.
  • Session‑scoped safety controls and budgeting via optional session ID.
  • Machine‑readable capabilities manifest available at .well-known/mcp-manifest.json.

Use cases of Muntuai Mcp

  • List and manage email outreach campaigns from an AI assistant.
  • Import leads, create senders, and verify domains programmatically.
  • Subscribe to workspace webhook events for real‑time monitoring.
  • Automate campaign lifecycle: upload, enrich, draft, send, pause, resume.

FAQ from Muntuai Mcp

How do I get an agent key?

Log in to app.muntuai.com, go to Settings → Agent Keys, create a key, and choose the autonomy level (observer, copilot, or autonomous). The raw key is shown once in the format mnt_<keyId>.<secret>.

What are the autonomy levels and what do they allow?

Observer allows reading all data (campaigns, leads, domains, etc.). Copilot adds create and modify actions for most resources. Autonomous includes everything above plus pause, resume, and delete operations.

What runtime or dependencies are needed?

No Muntu‑specific runtime is required. Clients connect over HTTP using any standard MCP client (Claude Desktop, Cursor) or a Python script with the requests library. The server is hosted by MuntuAI at https://api.muntuai.com/api/mcp.

How is authentication handled?

Authentication uses a Bearer token in the Authorization header. The token is the raw agent key (e.g., mnt_<keyId>.<secret>). It should be stored in an environment variable or secret manager and never committed to source control.

What transports and protocols does Muntuai Mcp support?

It uses the Model Context Protocol over HTTP, with Streamable HTTP transport (text/event-stream or application/json). The server protocol revision is 2025-03-26.

评论

其他 分类下的更多 MCP 服务器