channex-mcp
Enables interaction with the Channex.io API for managing properties, room types, rate plans, availability, rates, inventory, and channel connections through natural language.
README
Channex MCP
A self-improving Model Context Protocol (MCP) server for interacting with the Channex.io API.
⚠️ Important Disclaimer
This is NOT an official Channex.io project. This is an independent, open-source implementation of an MCP server that interfaces with the Channex.io API. It is not affiliated with, endorsed by, or supported by Channex.io. Use at your own risk.
🚀 Quick Start with Claude Code
See CLAUDE_CODE_SETUP.md for detailed instructions on adding this MCP to Claude Code.
🔧 MCP Configuration
For Cursor/Claude Desktop
Add the following to your ~/.cursor/mcp.json or Claude Desktop configuration:
{
"mcpServers": {
"channex": {
"command": "npx",
"args": ["--prefix", "/path/to/channex-mcp", "channex-mcp"],
"env": {
"MCP_MODE": "mcp",
"CHANNEX_API_KEY": "your-api-key-here",
"CHANNEX_BASE_URL": "https://app.channex.io/api/v1/"
}
}
}
}
Important: Replace /path/to/channex-mcp with the absolute path to your channex-mcp directory and add your actual Channex API key.
Common Issues
- ES Module errors: The project uses ES modules. The configuration above uses
npxto handle module loading correctly. - Server not starting: Ensure you've run
npm installandnpm run buildin the channex-mcp directory first.
Features
- ✨ Complete CRUD operations for Properties, Room Types, Rate Plans
- 📊 ARI (Availability, Rates, Inventory) management
- 🔄 Self-improving architecture with recursive commands
- 🧪 Built-in testing framework
- 📝 Auto-generated documentation
- 🔒 Secure API key management
Installation
npm install
Configuration
- Copy
.env.exampleto.env - Add your Channex API key
cp .env.example .env
Usage
Running the MCP Server
npm run dev
Self-Improvement Commands
Generate new endpoints:
npm run command generate-endpoint -- bookings list,get,create
Run tests:
npm run command test-all
Update documentation:
npm run command update-docs
Improve types from API responses:
npm run command improve-types -- properties samples/properties.json
Available Tools
The MCP server exposes the following tools:
Properties
channex_list_properties- List all propertieschannex_get_property- Get property by IDchannex_create_property- Create new propertychannex_update_property- Update propertychannex_delete_property- Delete property
Room Types
channex_list_room_types- List room typeschannex_get_room_type- Get room type by IDchannex_create_room_type- Create room type
Rate Plans
channex_list_rate_plans- List rate planschannex_get_rate_plan- Get rate plan by IDchannex_create_rate_plan- Create rate plan
ARI (Availability, Rates, Inventory)
channex_get_availability- Get availability per room typechannex_get_restrictions- Get restrictions per rate planchannex_update_ari- Update availability, rates, and restrictions
Channels (OTA Connections)
channex_test_channel_api- Test channel API accesschannex_check_existing_connection- Check for existing channel connectionschannex_list_channels- List all channel connections (supports field filtering)channex_get_channel_by_code- Get channels by code (optimized for specific channels)channex_get_channel- Get channel detailschannex_create_channel- Create new channel (e.g., Airbnb)channex_update_channel- Update channel settings (now supports property_ids)channex_delete_channel- Delete channel connectionchannex_get_channel_mappings- Get listing-to-rate-plan mappingschannex_update_channel_mapping- Map channel listings to rate planschannex_get_airbnb_listings- Get Airbnb-specific listingschannex_update_airbnb_listing- Update Airbnb pricing/availability
Development
Project Structure
channex-mcp/
├── src/
│ ├── index.ts # MCP server entry point
│ ├── api/
│ │ └── client.ts # Channex API client
│ ├── resources/ # Resource handlers
│ │ ├── properties.ts
│ │ ├── room-types.ts
│ │ ├── rate-plans.ts
│ │ ├── ari.ts
│ │ └── channels.ts
│ └── types/ # TypeScript definitions
├── .claude/
│ └── commands/ # Self-improvement scripts
└── CLAUDE.MD # Claude Code documentation
Adding New Features
- Use the
generate-endpointcommand to scaffold new resources - Add TypeScript types in
src/types/index.ts - Implement handlers in the MCP server
- Update documentation using
update-docs
Recent Updates (Jan 2025)
Response Size Optimization
- Added field filtering support to reduce response sizes
- Implemented response truncation for large objects
- Created optimized
channex_get_channel_by_codeendpoint - Fixed pagination parameter formatting
Channel Management Enhancement
- Added
property_idssupport tochannex_update_channel - Enables adding/removing properties from existing channels
- Essential for managing multi-property OTA connections
Contributing
We welcome contributions! Please see our Contributing Guidelines for details.
Important: This is an unofficial project. Contributors must:
- Test against real Channex APIs (no mocks)
- Respect Channex.io's Terms of Service
- Never commit API keys or credentials
License
This project is licensed under the MIT License - see the LICENSE file for details.
Disclaimer
This project is not affiliated with Channex.io. The Channex name and API are property of their respective owners. This is an independent project that provides an MCP interface to interact with the public Channex API.
Support
For issues and questions, please open a GitHub issue.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。