ebook
A MCP server that supports mainstream eBook formats including EPUB, PDF and more. Simplify your eBook user experience with LLM.
claude mcp add --transport stdio onebirdrocks-ebook-mcp npx -y onebirdrocks-ebook-mcp
How to use
Ebook-MCP provides a standardized Model Context Protocol (MCP) server for processing electronic books, focusing on EPUB and PDF formats. It enables natural language conversations with your digital library, integrates with AI-powered tools, and offers capabilities such as smart library management, interactive reading, active learning support, and content navigation. You can query your e-books, request summaries, extract key points from chapters, generate quizzes, and navigate to sections or topics using natural language prompts. This makes it easier to work with your books inside AI-enabled IDEs or conversational agents. The server is designed to bridge traditional e-books with AI assistants, so you can ask questions like “Show me chapters about prompt engineering” or “Summarize the key points from Chapter 5,” and receive guided, contextual responses.
The quick demo emphasizes a workflow with DeepSeek and examples that integrate with Claude/OpenAI through the mcp-client-example folder. Typical interactions include locating EPUB files, querying chapters, and generating learning tasks, all backed by the Ebook-MCP APIs.
How to install
Prerequisites:
- Node.js installed on your system (recommended latest LTS)
- Internet access to fetch the MCP package from npm/npx
Installation steps:
- Ensure you have npm installed and accessible in your shell.
- Install or run the MCP server via NPX (this example uses an npm package name assumed from the repository naming):
npm install -g onebirdrocks-ebook-mcp # if you prefer a global install
or rely on npx for a one-off run as shown in mcp_config
- Run the MCP server (per the mcp_config, using npx): npx -y onebirdrocks-ebook-mcp
- Verify the server starts and exposes its MCP API endpoints. If needed, configure environment variables for authentication, logging, or storage paths as described in the project documentation.
- (Optional) Integrate with your preferred MCP client or AI assistant by pointing it at the running server’s MCP endpoints.
Additional notes
Tips and considerations:
- The server focuses on EPUB and PDF processing; ensure your library paths and file access permissions allow reading these formats.
- If you encounter permission or network issues when using NPX or npm install, check your npm config and try running with elevated privileges or a clean npm cache.
- Common environment variables to consider adding include:
- VAR_API_KEY: for authentication with any integrated services
- LOG_LEVEL: e.g., debug, info, warn, error
- STORAGE_PATH: where to store processed or cached book data
- When troubleshooting, consult the mcp-client-example folder for example prompts and client usage patterns that show how to structure queries and interpret responses.
- If switching to a Docker or Python-based deployment later, be sure to adapt the mcp_config to the corresponding command and arguments.
Related MCP Servers
python -client
支持查询主流agent框架技术文档的MCP server(支持stdio和sse两种传输协议), 支持 langchain、llama-index、autogen、agno、openai-agents-sdk、mcp-doc、camel-ai 和 crew-ai
infobus
Model Context Protocol server enabling AI assistants to access transit information through standardized interfaces
mcp-gateway
MCP Gateway and Registry
mcp -email
一个基于 MCP (Model Context Protocol) 的邮件服务,支持 LLM 发送带附件的电子邮件及在指定目录中搜索文件。提供安全的 SMTP 传输、多收件人支持和附件模式匹配搜索功能,适用于 Gmail、Outlook、Yahoo、QQ 邮箱和网易 126 邮箱等主流邮箱服务。
pentestMCP
pentestMCP: AI-Powered Penetration Testing via MCP, an MCP designed for penetration testers.
MCPSecBench
MCPSecBench: A Systematic Security Benchmark and Playground for Testing Model Context Protocols