WeChat-MCP

Overview

WeChat-MCP enables LLM assistants to interact with WeChat programmatically by using macOS Accessibility API and screen capture. It provides tools for fetching recent messages, sending automated replies, publishing Moments posts, adding contacts, and smart chat search. The server includes 5 specialized Claude Code sub-agents for advanced WeChat automation tasks like summarizing chat history, auto-generating replies, searching messages, monitoring multiple chats, and analyzing communication patterns.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• fetch_messages_by_chat

  Get recent messages from a chat

  Try:Get recent messages from the John Doe chatTry:Show me the last 10 messages from our group
• reply_to_messages_by_chat

  Send a reply to a chat

  Try:Reply to the last message with 'Thanks for the update!'Try:Send a message to Jane asking about the meeting
• add_contact_by_wechat_id

  Add a new contact using a WeChat ID and send a friend request

  Try:Add contact with WeChat ID 'john123' with standard privacy settingsTry:Send a friend request to user 'work_contact'
• publish_moment_without_media

  Publish a text-only Moments post with optional draft-only mode

  Try:Publish a moment saying 'Great day at the conference!'Try:Create a draft moment about my vacation without posting it

Comparable tools

Installation

# Install via pip
pip install wechat-mcp-server

# Setup with Claude Desktop
{
  "mcpServers": {
    "wechat-mcp": {
      "type": "stdio",
      "command": "wechat-mcp"
    }
  }
}

FAQ

Which platforms are supported?

Currently only macOS with WeChat for Mac installed and running.

How to set up Accessibility permissions?

Go to System Settings → Privacy & Security → Accessibility, add your terminal app, and ensure WeChat is running before using the server.

Compare WeChat-MCP with