MCP 服务器

mcp-pdf-tools

Okay, here's the translation of the English text into Chinese, focusing on clarity and common usage in a technical context: **使用 PyPDF2 的 mcp 来实现:** * **合并 PDF 文件 (merge-pdfs)** * **提取页面 (extract-pages)** * **搜索 PDF 文件 (search-pdfs)** * **按指定顺序合并 PDF 文件 (merge-pdfs-ordered)** (This emphasizes the user-specified order) * **查找相关 PDF 文件 (find-related-pdfs)** (使用正则表达式从文本中提取信息来查找相关 PDF 文件) (This adds context about using regex) **Explanation of Choices:** * **mcp:** I've left "mcp" as is, assuming it's a specific tool or project name. If it's an acronym, you might want to spell it out in Chinese if you know the full name. * **合并 (hébìng):** The most common and direct translation for "merge." * **提取 (tíqǔ):** The standard translation for "extract." * **搜索 (sōusuǒ):** The standard translation for "search." * **页面 (yèmiàn):** The standard translation for "page." * **文件 (wénjiàn):** The standard translation for "file." * **按指定顺序 (àn zhǐdìng shùnxù):** This translates to "according to specified order" and is clearer than a more literal translation. * **查找相关 (cházhǎo xiāngguān):** This translates to "find related." * **正则表达式 (zhèngzé biǎodáshì):** The standard translation for "regular expression." * **从文本中提取信息 (cóng wénběn zhōng tíqǔ xìnxī):** This translates to "extract information from text." The added context in the last bullet point clarifies the method used for finding related PDFs. This makes the translation more informative.

Tools

merge-pdfs

Merge multiple PDF files into a single PDF

extract-pages

Extract specific pages from a PDF file

search-pdfs

Search for PDF files in a directory with optional pattern matching

merge-pdfs-ordered

Merge PDFs in a specific order based on patterns or exact names

find-related-pdfs

Find a PDF and then search for related PDFs based on its content, including common substring patterns

README

正在进行中 - 请谨慎使用 - Windows:

MCP PDF 工具服务器

一个提供 PDF 操作工具的 MCP (模型上下文协议) 服务器。该服务器允许 LLM 通过模型上下文协议执行诸如合并 PDF 和提取页面等操作。

功能

将多个 PDF 文件合并为一个 PDF
按照用户指定的顺序将多个 PDF 文件合并为一个 PDF
从 PDF 文件中提取特定页面
搜索 PDF 文件系统搜索或 Everything 搜索比此功能更好
基于文本提取和来自目标输入 PDF 的正则表达式模式匹配，查找（和合并）相关的 PDF

安装

克隆此存储库

cd mcp-pdf-tools

# 创建并激活虚拟环境
uv venv
.venv\Scripts\activate

# 安装包
uv pip install -e .

与 Claude Desktop 一起使用

将其添加到您的 Claude Desktop 配置文件 (claude_desktop_config.json) 中：

{
    "mcpServers": {
        "pdf-tools": {
            "command": "uv",
            "args": [
                "--directory",
                "PATH_TO\\mcp-pdf-tools",
                "run",
                "pdf-tools"
            ]
        }
    }
}