File 3f9c1f70

MCP文档知识

提供方 未标注

文档知识 SSE · 服务端推送 未标注 社区 / 第三方
查看接入方式

全面解析FILE CONVERTER MCPMCP Server的核心功能、安装配置和实用案例。作为顶级Model Context Protocol服务器,FILE CONVERTER MCP能让AI助手访问实时数据、执行操作,为您提供更智能的工作体验和自动化解决方案。

一句话选型结论

按资料字段整理

File 3f9c1f70 属于文档知识类 MCP 服务器,SSE · 服务端推送,鉴权方式未标注,有可直接复制的远程端点配置。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类文档知识 传输方式SSE · 服务端推送 鉴权未标注 部署方式远程部署
GitHub 星标 未标注 仓库热度,无公开仓库时显示未标注
工具数 未标注 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

文档知识
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量未标注暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

安装与配置

SSE · 服务端推送 未标注
早期远程传输方式,正在被 HTTP 取代
pip install -r requirements.txt

这条命令只是入口,具体启动参数请按官方文档补齐,不要直接照搬到别的服务器。

  1. 1
    准备运行环境

    确认你的客户端支持远程端点,并拿到服务地址。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后随便调一个只读工具,确认能返回结果,再接入正式流程。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
文档检索问答

在你的文档库里按问题找答案。

知识库读写

把新资料写进知识库,随用随取。

内容整理

把零散资料归纳成结构化笔记。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有现成的安装命令,不用自己找入口,照着跑就能起服务。

先想清楚的地方

  • 远程端点依赖网络与服务方可用性,注意密钥与调用额度。
  • 资料没标注工具数量,实际能力要装好后确认。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在文档知识场景里让模型直接调用外部能力的人。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
File 3f9c1f70当前 未标注 SSE · 服务端推送 远程部署 有 2026-05-17
dify 未标注 stdio · 本地进程 本地部署 有 2026-04-11
open-webui 未标注 stdio · 本地进程 本地部署 有 2026-04-11
JeecgBoot 未标注 stdio · 本地进程 本地部署 有 2026-04-11
Cherry Studio 未标注 未标注 未标注 无 2026-05-17
mempalace 未标注 stdio · 本地进程 本地部署 有 2026-04-11

服务器档案

文档知识
字段来自项目仓库与公开索引,中文说明由本站标注
提供方未标注上游标注的维护方 未标注
分类文档知识本站按上游分类展示
连接方式SSE · 服务端推送早期远程传输方式,正在被 HTTP 取代
鉴权方式未标注远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · Python本地安装时用到的包标识与运行时
部署类型远程部署上游给出的部署形态
最近核验2026-05-17上游最后核验时间

详细介绍

8.6 KB
来自项目仓库与上游目录的原文,未做删改

Pandoc MCP Server

A Python-based MCP (Model Context Protocol) server that provides powerful document conversion capabilities via Pandoc. This server allows AI agents (like Claude via LangChain/LangGraph) to request file conversions between various formats such as Markdown, DOCX, HTML, PDF, EPUB, and many more.

This project uses:

  • FastMCP: A Python library for easily creating MCP servers.
  • pypandoc: A Python wrapper around the Pandoc command-line tool.
  • Pandoc: The universal document converter.
  • (Optional) Docker: For containerized deployment, bundling all dependencies (Python, Pandoc, LaTeX).

Features

  • Exposes a single MCP tool: convert_document.
  • Supports a wide range of input and output formats handled by Pandoc.
  • Allows specifying input format (if auto-detection fails) and output format.
  • Supports passing extra command-line arguments to Pandoc for advanced control (e.g., Table of Contents, PDF margins, standalone files).
  • Includes Docker configuration (Dockerfile) for creating a self-contained server environment including Pandoc and necessary LaTeX components for PDF generation.
  • Designed for integration with MCP clients, particularly LangChain/LangGraph agents.

Exposed MCP Tool

convert_document

Converts a document from one format to another using Pandoc.

Arguments:

  • input_file_path (str, required): The path accessible by the server to the input document file. If running in Docker with a volume mount, this should be the path inside the container (e.g., /data/my_doc.docx).
  • output_file_path (str, required): The path accessible by the server where the converted output file should be saved. If running in Docker, this should be the path inside the container (e.g., /data/my_output.pdf). The directory will be created if it doesn't exist within the server's accessible filesystem.
  • to_format (str, required): The target format for the conversion (e.g., 'markdown', 'docx', 'pdf', 'html', 'rst', 'epub'). See Pandoc documentation for a full list (--list-output-formats).
  • from_format (str, optional): The format of the input file. If None, pandoc will try to guess from the file extension. Specify if the extension is ambiguous or missing (e.g., 'md', 'docx', 'html'). Defaults to None.
  • extra_args (List[str], optional): A list of additional command-line arguments to pass directly to pandoc (e.g., ['--toc'], ['-V', 'geometry:margin=1.5cm'], ['--standalone']). Defaults to None.

Returns:

  • (str): A message indicating success (e.g., "Successfully converted document to '/data/my_output.pdf'") or an error message (e.g., "Error: Input file not found...", "Error during conversion: Pandoc died...").

Setup and Running

You can run this server either locally (requires manual installation of dependencies) or using the provided Docker configuration (recommended for ease of use and deployment).

Installing via Smithery

To install Pandoc Document Converter for Claude Desktop automatically via Smithery:```bash
npx -y @smithery/cli install @MaitreyaM/file-converter-mcp --client claude


### Option 1: Running with Docker (Recommended)

This method bundles Python, Pandoc, LaTeX, and required libraries into a container. **You only need Docker Desktop installed locally.**

1.  **Install Docker:** Download and install Docker Desktop for your operating system. Start Docker Desktop.
2.  **Clone Repository:** Get the project files:```bash
    git clone https://github.com/your-username/pandoc-mcp-server.git # Replace with your repo URL
    cd pandoc-mcp-server
    ```
3.  **Build the Docker Image:** This command builds the image using the `Dockerfile`. It installs Pandoc, a capable TeX Live distribution (for PDF support), and Python dependencies inside the image. This step might take several minutes the first time.```bash
    docker build -t pandoc-converter-server .
    ```
4.  **Run the Container:** This starts the server inside the container.
    *   **Choose a directory on your host machine** to share with the container for input/output files (e.g., the current project directory).
    *   Run the container, mapping the host directory to `/data` inside the container and mapping port 8000. **Replace `/path/to/your/local/project` with the actual absolute path to the project directory on your machine.**```bash
    # Example using the current directory (.) as the host path:
    docker run -it --rm -p 8000:8000 -v "$(pwd)":/data pandoc-converter-server

    # Or using an absolute path (replace):
    # docker run -it --rm -p 8000:8000 -v "/path/to/your/local/project":/data pandoc-converter-server
    ```
    *   `-it`: Runs interactively (shows logs, allows Ctrl+C).
    *   `--rm`: Removes the container when stopped.
    *   `-p 8000:8000`: Maps port 8000 on your host to port 8000 in the container.
    *   `-v "$(pwd)":/data`: Mounts the current working directory on your host to `/data` inside the container. Files placed in your local project directory will appear in `/data` inside the container, and files saved to `/data` by the server will appear in your local project directory.
    *   `pandoc-converter-server`: The name of the image you built.
5.  **Server is Running:** You should see logs indicating the server started and is listening on SSE (`http://0.0.0.0:8000`). It's ready to accept connections from your MCP client (like the LangChain agent).
6.  **Connecting from Client:** Configure your MCP client (e.g., `MultiServerMCPClient`) to connect to `http://127.0.0.1:8000/sse` with `transport: "sse"`.
7.  **Using the Tool:** When interacting with your agent/client, refer to files using their path *inside the container*, prefixed with `/data/`. For example: `convert /data/my_input.docx to pdf at /data/my_output.pdf`. The output file will appear in your local project directory due to the volume mapping.

### Option 2: Running Locally (Manual Dependency Installation)

This requires you to install Python, Pandoc, and a LaTeX distribution directly onto your host machine.

1.  **Install Python:** Ensure you have Python >= 3.10 installed.
2.  **Install Pandoc:** Install the Pandoc command-line tool for your OS. Follow instructions at pandoc.org/installing.html. Verify by running `pandoc --version` in a new terminal.
3.  **Install LaTeX:** For PDF generation, install a TeX distribution.
    *   **macOS:** `brew install --cask mactex-no-gui` (Recommended via Homebrew)
    *   **Debian/Ubuntu:** `sudo apt-get update && sudo apt-get install texlive-latex-base texlive-fonts-recommended texlive-latex-extra texlive-fonts-extra` (or `texlive-full` for everything, but large).
    *   **Windows:** Install MiKTeX or TeX Live. Ensure the `bin` directory containing `pdflatex.exe` is added to your system's PATH.
    *   Verify by running `pdflatex --version` in a new terminal.
4.  **Clone Repository:**```bash
    git clone https://github.com/your-username/pandoc-mcp-server.git # Replace with your repo URL
    cd pandoc-mcp-server
    ```
5.  **Create Virtual Environment (Recommended):**```bash
    python -m venv venv
    source venv/bin/activate # Linux/macOS
    # venv\Scripts\activate # Windows
    ```
    *(Or use Conda: `conda create --name pandoc-env python=3.11 && conda activate pandoc-env`)*
6.  **Install Python Dependencies:**```bash
    pip install -r requirements.txt
    ```
7.  **Run the Server:**```bash
    python pandoc_mcp_server.py
    ```
8.  **Server is Running:** It will listen on `http://127.0.0.1:8000/sse`.
9.  **Connecting from Client:** Configure your MCP client to connect to `http://127.0.0.1:8000/sse`.
10. **Using the Tool:** Refer to files using their regular paths on your local machine (e.g., `convert my_input.docx to pdf at my_output.pdf`, assuming files are in the same directory, or use absolute paths).

## Example Agent Interaction (Running Server in Docker)

Assuming the server container is running with the volume mount:```
You: convert /data/report.md to pdf

Agent: Thinking...
[Agent calls convert_document tool with input='/data/report.md', output='/data/report.pdf', to='pdf']
Agent: Successfully converted document to '/data/report.pdf'
[The bot may then attempt to upload report.pdf from the local project directory]

Files

  • pandoc_mcp_server.py: The main Python script for the MCP server.
  • Dockerfile: Instructions for building the Docker container image.
  • requirements.txt: Python dependencies needed inside the Docker container (or local venv).
  • .gitignore: Specifies intentionally untracked files for Git.
  • README.md: This file.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request or open an Issue.

常见问题

7 问
答案由本站依据条目字段整理,不引用上游原文
File 3f9c1f70 怎么安装?

在客户端按下面的安装命令接入:pip install -r requirements.txt。装完在客户端配置里引用同一个包或端点。

File 3f9c1f70 怎么在客户端里配置?

它的传输方式未标注,在客户端 MCP 配置里按官方文档填写启动命令和参数,保存后重启客户端生效。

File 3f9c1f70 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

File 3f9c1f70 需要 API Key 吗?

资料没有标注鉴权方式,是否需要密钥取决于服务本身,接入前按官方文档确认。

File 3f9c1f70 能提供哪些能力?

资料显示能力标签包括 文档转换、Python、SSE、文档处理。具体工具清单以官方文档为准。

File 3f9c1f70 需要什么运行环境?

运行时为 Python;属于远程端点,不需要在本地装运行时。

File 3f9c1f70 是本地还是远程?

它是远程端点,连服务地址即可。传输方式为SSE · 服务端推送。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 2026-05-17。