file-convert-mcp
A Model Context Protocol (MCP) server for converting files between different formats
Documentation
File Convert MCP Server
A Model Context Protocol (MCP) server for converting files between various formats.
File Convert is an MCP server designed to handle the conversion of a wide array of file formats, including images, office documents, audio, video, text, and data files. It aims to provide seamless transformation into popular formats such as PDF, JPG, MP4, and HTML, among others. This server offers a set of powerful tools to convert diverse file types into easily accessible and widely compatible formats like PDF, JPG, PNG, TXT.
✨ Features
- Intelligent File Type Detection: Employs a combination of AI, TrID, and magic bytes for accurate file type identification.
- Versatile File Conversion: Supports conversion between numerous file types and the following popular formats:
🛠️ Requirements
1. Python: Version 3.12 or higher is required.
2. uv Tool: Install the Python uv tool for efficient dependency management and easy execution of the MCP server.
3. API Key (Development): As of May 2025, the MCP is in its testing phase, and no API key is currently required for development. For future production use, please contact the Convert.Guru team to obtain an API key.
🚀 Usage with Desktop App
To integrate this server with your desktop application, add the following configuration to your app's server settings. If necessary, adjust the absolute path to the `uv`/`uvx` executables (or `uv.exe`/`uvx.exe` on Windows).
Using `uvx` (recommended):
{
"mcpServers": {
"file-convert": {
"command": "uvx",
"args": ["--from", "git+https://github.com/convertguru/file-convert-mcp.git", "file-convert-mcp"],
"env": {
"CONVERT_GURU_API_KEY": "your_api_key_here"
}
}
}
}If you have cloned the repository locally (adjust paths as needed):
{
"mcpServers": {
"file-convert": {
"command": "/home/User/.local/bin/uv",
"args": ["--directory", "/home/User/file-convert-mcp/src/file_convert_mcp", "run", "server.py"],
"env": {
"CONVERT_GURU_API_KEY": "your_api_key_here"
}
}
}
}Using local `uvx`:
{
"mcpServers": {
"file-convert": {
"command": "uvx",
"args": ["--from", "/home/User/file-convert-mcp", "file-convert-mcp"],
"env": {
"CONVERT_GURU_API_KEY": "your_api_key_here"
}
}
}
}Environment Variables
The MCP server supports the following environment variables:
- `CONVERT_GURU_API_KEY`: Your Convert.Guru API key (optional for development)
- `TRANSPORT`: Transport method for the MCP server
- `"stdio"` (default): Standard input/output transport
- `"http"`: HTTP transport (streamable) with REST API endpoints
- `PORT`: Port number for HTTP transport (default: 8000, only used when `TRANSPORT=http`)
When using HTTP transport, the server provides additional endpoints:
- `GET /health` - Health check endpoint
- `GET /tools` - List of available MCP tools
Also, see this guide on how to set up MCP tools in Claude Desktop.
Installing via Smithery
To install file-convert-mcp for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @convertguru/file-convert-mcp --client claude🛠️ Development
Get started with local development by following these steps:
1. Clone the repository:
git clone https://github.com/convertguru/file-convert-mcp.git2. Fetch and cache Python dependencies (optional, `uvx` handles this automatically):
cd file-convert-mcp
uv sync
cd ..3. Create .env file with your (optional for now) API key:
echo "CONVERT_GURU_API_KEY=your_api_key_here" > file-convert-mcp/.env4. Run the MCP server locally using various `uv` commands:
# Using uvx with .env file from the local directory
cd file-convert-mcp
UV_ENV_FILE=.env uvx --from ./file-convert-mcp file-convert-mcp
# OR using uv directly to run the server script
uv run file-convert-mcp/src/file_convert_mcp/server.py
# OR specifying the directory for uv
uv --directory file-convert-mcp/src/file_convert_mcp run server.py
# OR navigating into the server directory
cd file-convert-mcp/src/file_convert_mcp
uv run server.py
# OR using uvx to fetch the core from the GitHub repository + local .env file
UV_ENV_FILE=.env uvx --from git+https://github.com/convertguru/file-convert-mcp.git file-convert-mcp
# OR run with HTTP transport on custom port
TRANSPORT=http PORT=9000 uv run file-convert-mcp/src/file_convert_mcp/server.py
# OR run with stdio transport (default)
TRANSPORT=stdio uv run file-convert-mcp/src/file_convert_mcp/server.py5. Create .env file with your configuration (optional for now):
# Basic configuration
echo "CONVERT_GURU_API_KEY=your_api_key_here" > file-convert-mcp/.env
# OR with transport configuration
echo "CONVERT_GURU_API_KEY=your_api_key_here" > file-convert-mcp/.env
echo "TRANSPORT=http" >> file-convert-mcp/.env
echo "PORT=8000" >> file-convert-mcp/.env6. Modify the server logic if needed:
Edit the main server file located at `src/file_convert_mcp/server.py.`
7. Clearing the `uv` Cache (if needed):
If `uv` has cached an older version of the code in `~/.cache/uv`, you might need to clear the cache. Alternatively, use `uv` with the `-n` or `--no-cache` option to bypass it.
⚙️ Available Tools
The MCP server provides the following tools:
- `detect_file_type`: Analyzes the first 200 bytes of an uploaded file and uses the Convert.Guru API to determine its type.
- `convert_file`: Converts a given file to a specified output format. The desired file extension should be passed as the `ext_out` parameter.
📜 License
Visit us https://convert.guru
This project is licensed under the MIT License - see the LICENSE file for complete details.
Frequently asked questions
What is file-convert-mcp?
file-convert-mcp is A Model Context Protocol (MCP) server for converting files between different formats
How do I install file-convert-mcp?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is file-convert-mcp open source?
Yes — it is hosted on GitHub at https://github.com/convertguru/file-convert-mcp and has 5 stars.
Related MCP tools
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
Automate browser based workflows with AI
Hindsight: Agent Memory That Learns
A privacy-first app that strips AI watermarks from content you own.
Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.
The power of Claude Code / GeminiCLI / CodexCLI + [Gemini / OpenAI / OpenRouter / Azure / Grok / Ollama / Custom Model / All Of The Above] working as one.
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP