ebook-mcp

概述

Ebook-MCP 是一个强大的模型上下文协议服务器，旨在改变用户与数字图书的交互方式。它提供标准化 API，实现 LLM 应用与电子书处理功能的无缝集成。该服务器支持主流电子书格式，包括 EPUB 和 PDF，使用户能够通过自然语言查询执行图书馆管理、内容提取和交互式阅读等任务。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• get_all_epub_filesget_all_epub_files(path: str) -> List[str]

  Get all EPUB files in the specified directory

  试试:Show me all EPUB files in my downloads folder试试:Find all EPUB books about GenAI in my library
• get_metadataget_metadata(epub_path: str) -> Dict[str, Union[str, List[str]]]

  Get metadata from an EPUB file

  试试:Get the title and author of this book试试:Show me publication details for this EPUB
• get_tocget_toc(epub_path: str) -> List[Tuple[str, str]]

  Get table of contents from an EPUB file

  试试:Show me the table of contents试试:What chapters are in this book?
• get_chapter_markdownget_chapter_markdown(epub_path: str, chapter_id: str) -> str

  Get chapter content in Markdown format

  试试:Get Chapter 3 content in Markdown试试:Show me the introduction chapter
• get_all_pdf_filesget_all_pdf_files(path: str) -> List[str]

  Get all PDF files in the specified directory

  试试:List all PDF files in my documents folder试试:Find PDF books about machine learning
• get_pdf_metadataget_pdf_metadata(pdf_path: str) -> Dict[str, Union[str, List[str]]]

  Get metadata from a PDF file

  试试:Get the author and creation date of this PDF试试:Show me publication information for this document
• get_pdf_tocget_pdf_toc(pdf_path: str) -> List[Tuple[str, int]]

  Get table of contents from a PDF file

  试试:Show me the table of contents for this PDF试试:What chapters are in this document?
• get_pdf_page_textget_pdf_page_text(pdf_path: str, page_number: int) -> str

  Get plain text content from a specific page

  试试:Get the text from page 10试试:Show me page 5 content
• get_pdf_page_markdownget_pdf_page_markdown(pdf_path: str, page_number: int) -> str

  Get Markdown formatted content from a specific page

  试试:Get page 10 in Markdown format试试:Convert page 3 to Markdown
• get_pdf_chapter_contentget_pdf_chapter_content(pdf_path: str, chapter_title: str) -> Tuple[str, List[int]]

  Get chapter content and corresponding page numbers by chapter title

  试试:Get content for Chapter 1试试:Show me the section about RAG

可对比工具

安装

# 克隆仓库
git clone https://github.com/onebirdrocks/ebook-mcp.git
cd ebook-mcp

# 使用 uv 安装依赖
uv pip install -e .

# 在开发模式下启动 MCP 服务器
uv run mcp dev src/ebook_mcp/main.py

# 或在生产模式下启动
uv run src/ebook_mcp/main.py

{
  "ebook-mcp": {
    "command": "uv",
    "args": [
      "--directory",
      "/path/to/ebook-mcp/src/ebook_mcp/",
      "run",
      "main.py"
    ]
  }
}

FAQ

这个服务器支持哪些电子书格式？

Ebook-MCP 目前支持 EPUB 和 PDF 格式，具有元数据提取、目录检索和内容提取等功能。

如何将此 MCP 服务器与 AI 应用集成？

该服务器可通过模型上下文协议与 AI 应用集成。提供了 Claude Desktop、Cursor 和其他兼容 AI 工具的安装说明。

ebook-mcp 对比