fsmw/kimi-writer-mcp

Kimi Writer MCP Server 🚀

A complete Model Context Protocol (MCP) server that exposes Kimi Writer's creative writing capabilities to other AI models. Transform your AI writing workflow with professional document generation and structured prompts.

✨ Features

🔧 Writing Tools (8 tools)

• create_project - Create organized writing projects
• write_file - Write markdown with create/append/overwrite modes
• get_project_info - Get active project information
• list_project_files - List project files with metadata
• read_file - Read file contents from active project
• get_file_stats - Get file statistics (size, dates, etc.)
• create_writing_template - Generate templates for novels, short stories, books, poetry

📝 Structured Prompts (3 prompts)

• write_novel - Complete novel writing prompts with genre-specific guidance
• write_short_story - Short story prompts with tone and length options
• write_nonfiction_book - Non-fiction book prompts with audience targeting

📚 Document Generation (Optional - 2 tools)

• generate_pdf - Professional PDFs with TOC, typography, pagination
• generate_epub - Standards-compliant EPUBs with metadata and navigation

🚀 Quick Start

1. Clone the repository

git clone https://github.com/fsmw/kimi-writer-mcp.git
cd kimi-writer-mcp

2. Install dependencies

# Install MCP dependencies
pip install -r requirements.txt

# Install Kimi Writer dependencies
cd ../kimi-writer
pip install -r requirements.txt
cd ../kimi-writer-mcp

# For PDF/EPUB generation (optional)
pip install -r requirements-extended.txt

3. Run the server

python mcp-server.py

4. Test functionality

python test-client.py

🤖 Claude Desktop Integration

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "kimi-writer": {
      "command": "python",
      "args": ["/path/to/kimi-writer-mcp/mcp-server.py"],
      "cwd": "/path/to/kimi-writer-mcp"
    }
  }
}

💻 Usage Examples

Writing a Novel with PDF Export

# Create project and write content
await session.call_tool("create_project", {"project_name": "cyberpunk_novel"})
await session.call_tool("write_file", {
    "filename": "chapter_01.md",
    "content": "# Chapter 1\n\nIn the neon-lit streets...",
    "mode": "create"
})

# Generate PDF when complete
await session.call_tool("generate_pdf", {"output_filename": "complete_novel"})

Creating a Short Story Collection

# Get structured prompt
prompt = await session.get_prompt("write_short_story", {
    "theme": "AI and consciousness",
    "length": "medium",
    "tone": "philosophical"
})

# Write stories using the prompt
# ... your AI model generates content ...

📊 Project Status

✅ COMPLETE AND TESTED

• ✅ All 8 MCP tools functioning correctly
• ✅ Project creation and management working
• ✅ File operations (read, write, append, overwrite) tested
• ✅ Template generation for all writing types
• ✅ Structured prompts with specific guidance
• ✅ Professional documentation and examples
• ✅ Cross-platform compatibility

🔧 Architecture

• MCP Protocol Compliance - Universal AI model compatibility
• Modular Design - Clean separation of concerns
• Graceful Degradation - Optional features work without errors
• Professional Code Quality - Kimi Writer standards with comprehensive docstrings
• Unicode Support - International content handling
• Cross-Platform - Unix/Linux/Windows compatibility

📦 Dependencies

Core (Required)

• mcp>=1.0.0 - Model Context Protocol
• openai>=1.0.0 - OpenAI SDK (Kimi Writer compatibility)
• httpx>=0.24.0 - HTTP client
• python-dotenv>=1.0.0 - Environment management
• markdown>=3.5.1 - Markdown processing

Document Generation (Optional)

• weasyprint>=60.0 - PDF generation
• ebooklib>=0.18 - EPUB creation
• beautifulsoup4>=4.12.0 - HTML parsing
• python-frontmatter>=1.0.0 - Metadata handling

🧪 Testing

Run comprehensive test suites:

# Test all MCP features
python test-client.py

# Test document generation
python test-documents.py

# Check dependencies
python start.py --check

📚 Documentation

• - Complete documentation
• - Technical details
• - Usage examples
• - Project summary

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Run tests: python test-client.py
5. Submit a pull request

📄 License

MIT License - see file for details.

🌟 Credits

Created by Fernando San Martin with 🧡 Crush

Transforming Kimi Writer into a powerful, reusable writing tool that can be integrated with any MCP-compatible AI system.

🚀 Ready for Claude Desktop integration!