home / mcp / weread mcp server

Weread MCP Server

微信读书MCP

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "freestylefly-mcp-server-weread": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-server-weread"
      ],
      "env": {
        "CC_ID": "YOUR_ID",
        "CC_URL": "https://cc.chenge.ink",
        "CC_PASSWORD": "YOUR_PASSWORD",
        "WEREAD_COOKIE": "COOKIE_VALUE"
      }
    }
  }
}

This MCP server provides a dedicated bridge between WeChat Reading data and MCP-enabled language model clients. It exposes your bookshelf, notes, highlights, and popular reviews to compatible tools, enabling seamless querying, organization by chapters, and feeding structured data to your preferred LLM client.

How to use

You connect an MCP client to this server to access your WeChat Reading data. The server exposes four core capabilities: fetch your bookshelf, search your bookshelf for specific titles, retrieve notes and highlights organized by book chapters, and obtain popular reviews for a book. Use these tools through your MCP client to drive conversations, generate summaries, or prepare data-driven responses.

How to install

Prerequisites: install Node.js 16.x or higher and ensure you are signed into WeChat Reading with a valid cookie.

Option A: Run with npx (recommended for simplicity) — Claude Desktop integration example.

{
  "mcpServers": {
    "mcp-server-weread": {
      "command": "npx",
      "args": ["-y", "mcp-server-weread"],
      "env": {
        // Cookie Cloud approach is recommended
        "CC_URL": "https://cc.chenge.ink",
        "CC_ID": "您的ID",
        "CC_PASSWORD": "您的密码"
        // Alternatively, you can provide the WeChat Reading cookie directly via WEREAD_COOKIE
        // "WEREAD_COOKIE": "您的微信读书Cookie"
      }
    }
  }
}

Option B: Global install and run locally — you can install the MCP server globally and run it from any terminal.

npm install -g mcp-server-weread
```
```
{
  "mcpServers": {
    "mcp-server-weread": {
      "command": "mcp-server-weread",
      "env": {
        // Cookie Cloud approach is recommended
        "CC_URL": "https://cc.chenge.ink",
        "CC_ID": "您的ID",
        "CC_PASSWORD": "您的密码"
        // Alternatively, you can provide the WeChat Reading cookie directly via WEREAD_COOKIE
      }
    }
  }
}

Configuration and cookie management

To keep cookie credentials valid, you can use CookieCloud to auto-sync cookies. This avoids frequent cookie refreshes and keeps the MCP server authenticated.

CookieCloud setup steps (high level): install the browser extension and configure the server URL, then enable automatic synchronization. In MCP JSON, point to your CookieCloud URL and credentials as shown in the examples above.

If you prefer direct cookie usage, you can set WASREAD_COOKIE in the environment with your WeChat Reading cookie value. The system will try CookieCloud first and fall back to WEREAD_COOKIE if needed.

Examples of common actions

Fetch your bookshelf information to see all titles along with authors and categories.

Search your bookshelf by keywords to locate specific books.

Get notes and highlights for a book, organized by chapter, with filters for highlight style.

Retrieve hot reviews for a book, with pagination support.

Available tools

get_bookshelf

Fetches all books on the user’s WeChat Reading shelf, returning basic book information such as title, author, translator, and category.

search_books

Searches the user’s shelf by keyword with fuzzy and exact matching options, optionally returning detailed information and limiting results.

get_book_notes_and_highlights

Retrieves all notes and highlights for a specified book, organized by chapter, with optional filters by highlight style and structured data for LLM consumption.

get_book_best_reviews

Fetches popular reviews for a specified book, supports pagination, and includes rating, like count, and reviewer details.