Calibre

Exposes Calibre libraries to AI agents for search, metadata updates, and content access with granular permissions.

python

0

GitHub Stars

python

Language

3 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "facedeer-calibre_full_mcp_server": {
      "command": "python",
      "args": [
        "<full path>/calibre_full_mcp/src/server.py"
      ],
      "env": {
        "PYTHONPATH": "<full path>/calibre_full_mcp/src",
        "CALIBREMCP_CONFIGPATH": "<full path>/calibre_full_mcp/config.json"
      }
    }
  }
}

This MCP server bridges AI agents with your Calibre ebook libraries, enabling powerful search, metadata management, library maintenance, format conversion, direct reading, and granular permissions. It lets you expose your Calibre collections to agents as a dynamic knowledge base while allowing controlled updates when permissions permit.

How to use

You will connect an MCP client to the Calibre MCP Server to interact with your libraries. Use the client to perform advanced searches by metadata or full text, view and update book details, add or remove titles, convert formats on the fly, and read book content directly into the agent’s context for analysis or Q&A. Permissions are granular, so you can grant read access, restrict writes to specific fields, and control visibility per library.

How to install

Prerequisites you must meet before starting:

Place your configuration file and start the MCP server using the following steps. Replace the paths with your actual locations.

Step by step commands

# 1) Create the MCP configuration file at /path/to/config.json with the example configuration below.

{
  "libraries": {
    "default": {
      "path": "d:/ebooks/main_library",
      "description": "The primary research library containing technical manuals.",
      "default": true,
      "permissions": {
        "read": ["title", "authors", "tags", "rating", "comments"],
        "write": ["tags", "rating", "comments"],
        "delete": false,
        "convert": true
      },
      "import": {
        "allowed_paths": ["d:/downloads/ebook_imports"],
        "allow_delete_source": false
      },
      "export": {
        "allowed_paths": ["d:/ebook_exports"],
        "allow_overwrite_destination": false
      }
    }
  },
  "port": 8000,
  "enable_worker_logging": false,
  "expose_resources_via_tools": true,
  "log_level": "warning"
}

Step by step commands

Run the MCP server using the Python interpreter and specify the configuration path. Ensure the Python path includes the MCP source if required by your setup.

CALIBREMCP_CONFIGPATH=/path/to/config.json 
PYTHONPATH=/path/to/calibre_full_mcp/src 
python /path/to/calibre_full_mcp/src/server.py

Notes on running

The server exposes the MCP endpoint on the configured port (default 8000). Do not run the server against a library currently in use by the Calibre desktop application or other processes to prevent database corruption. You can modify the configuration to point to different libraries or adjust permissions as your needs evolve.