Huoshui File Converter
A secure MCP server for converting documents between Markdown, DOCX, HTML, PDF, and TXT formats within a sandboxed working directory.
README
Huoshui File Converter
A secure MCP (Model Context Protocol) server for document format conversion within a specified working directory.
Features
- š Sandbox Security: All operations restricted to a configured working directory
- š Format Support: Convert between Markdown, DOCX, HTML, PDF, and TXT
- š MCP Integration: Full MCP protocol support with prompts, resources, and tools
- āļø Flexible Configuration: CLI arguments, environment variables, or current directory
- š Smart Detection: Intelligent file format detection by content analysis
Quick Start
Installation
Option 1: From MCP Registry (Recommended)
This server is available in the Model Context Protocol Registry. Install it using your MCP client.
mcp-name: io.github.huoshuiai42/huoshui-file-converter
Option 2: Using uvx
uvx huoshui-file-converter
Option 3: Using pip
pip install huoshui-file-converter
Basic Usage
# Use current directory
uvx huoshui-file-converter
# Specify working directory (recommended)
uvx huoshui-file-converter --dir "/path/to/documents"
# Short form
uvx huoshui-file-converter -d "~/Documents"
MCP Client Configuration
For Claude Desktop or other MCP clients:
{
"mcpServers": {
"huoshui-converter": {
"command": "uvx",
"args": ["huoshui-file-converter", "--dir", "/Users/yourname/Documents"]
}
}
}
Configuration Options
Priority Order
- CLI Argument (highest priority):
--diror-d - Environment Variable:
HUOSHUI_WORKING_DIR - Smart Default: Documents folder if current directory is problematic
- Current Directory (fallback)
Examples
# CLI argument (best for MCP clients)
uvx huoshui-file-converter --dir "/project/docs"
# Environment variable
export HUOSHUI_WORKING_DIR="/project/docs"
uvx huoshui-file-converter
# Current directory fallback
cd /project/docs
uvx huoshui-file-converter
Supported Conversions
| From | To |
|---|---|
| Markdown | DOCX, HTML, PDF |
| DOCX | Markdown, HTML, PDF |
| HTML | Markdown, DOCX, PDF |
| TXT | Markdown, DOCX, HTML, PDF |
MCP Tools & Resources
Tools
convert_document: Convert files between formatsdetect_format: Intelligent format detection
Resources
file_list: Browse directory contents (optimized for large directories)limit: Control number of files shown (default: 100)supported_only: Show only convertible files
file_get: Get detailed file informationconversion_capability_list: List supported conversions
Prompts
role_and_rules: AI assistant behavior guidelines
Performance Features
- Fast Directory Listing: Extension-based format detection for large directories
- Smart File Limits: Default 100-file limit prevents UI freezing
- Large File Handling: Files >50MB are marked and handled specially
- Selective Display: Option to show only supported file formats
- Memory Efficient: Avoids reading file contents during directory browsing
Security Features
- Path Validation: Prevents directory traversal attacks
- Working Directory Restriction: All operations sandboxed to configured directory
- Startup Validation: Checks directory existence and permissions
- Relative Path Enforcement: Absolute paths are rejected
Command Line Options
$ uvx huoshui-file-converter --help
usage: huoshui-file-converter [-h] [--dir PATH] [--version]
Huoshui Document Converter - MCP Server for file conversion within a working directory
options:
-h, --help show this help message and exit
--dir PATH, -d PATH
Working directory for file operations (default: current directory or HUOSHUI_WORKING_DIR env var)
--version, -v show program's version number and exit
Examples:
uvx huoshui-file-converter # Use current directory
uvx huoshui-file-converter --dir /docs # Use specific directory
uvx huoshui-file-converter -d ./project # Use relative directory
Configuration Priority:
1. CLI argument (--dir/-d)
2. Environment variable (HUOSHUI_WORKING_DIR)
3. Current working directory
Error Handling
The server validates the working directory on startup:
ā
Working directory configured: /Users/name/Documents
š Source: CLI argument
Common errors and solutions:
| Error | Solution |
|---|---|
| Directory not found | Create directory or fix path |
| No write access | Check permissions (chmod on Unix) |
| Path outside sandbox | Use relative paths only |
Development
Requirements
- Python 3.8+
- pypandoc
- pandoc (system dependency)
- LaTeX (for PDF conversion)
Testing
# Test configuration
uvx huoshui-file-converter --dir "/tmp/test"
# Check startup messages
# ā
Working directory configured: /tmp/test
# š Source: CLI argument
Documentation
License
[Your license here]
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.