Oomol Connect MCP Server
Enables integration with Oomol Connect for executing tasks, managing blocks and packages, and uploading files. Supports task execution with real-time progress monitoring and intelligent polling across audio, video, and other processing workflows.
README
Oomol Connect MCP SDK
MCP (Model Context Protocol) SDK for Oomol Connect, enabling integration with Cherry Studio, VSCode, and other MCP-compatible clients.
Features
- ✅ Full MCP protocol support based on official
@modelcontextprotocol/sdk - ✅ 9 tools covering all Oomol Connect functionality
- ✅ Complete TypeScript type support
- ✅ File upload support (local paths and base64 encoding)
- ✅ Intelligent polling with exponential backoff
- ✅ Real-time progress monitoring
- ✅ Smart version filtering (returns latest version by default)
- ✅ Easy integration with Cherry Studio, VSCode, Claude Desktop
Installation
Global Installation
npm install -g oomol-connect-mcp-sdk-ts
Using npx (Recommended)
npx -y oomol-connect-mcp-sdk-ts
Local Development
git clone <repository-url>
cd oomol-connect-mcp-sdk-ts
npm install
npm run build
Configuration
Environment Variables
| Variable | Description | Required | Default |
|---|---|---|---|
OOMOL_CONNECT_BASE_URL |
API base URL | Yes | - |
OOMOL_CONNECT_API_TOKEN |
API token | Yes | - |
MCP_SERVER_NAME |
MCP server name | No | "oomol-connect" |
MCP_SERVER_VERSION |
MCP server version | No | package.json version |
OOMOL_CONNECT_DEFAULT_TIMEOUT |
Default timeout (ms) | No | 300000 (5 min) |
Example Configuration
# Required
export OOMOL_CONNECT_BASE_URL="http://localhost:3000/api"
export OOMOL_CONNECT_API_TOKEN="api-your-token-here"
# Optional
export OOMOL_CONNECT_DEFAULT_TIMEOUT="300000"
MCP Client Integration
Cherry Studio
Add to Cherry Studio's MCP configuration:
{
"mcpServers": {
"oomol-connect": {
"command": "npx",
"args": ["-y", "oomol-connect-mcp-sdk-ts"],
"env": {
"OOMOL_CONNECT_BASE_URL": "http://localhost:3000/api",
"OOMOL_CONNECT_API_TOKEN": "your-token-here"
}
}
}
}
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"oomol-connect": {
"command": "npx",
"args": ["-y", "oomol-connect-mcp-sdk-ts"],
"env": {
"OOMOL_CONNECT_BASE_URL": "http://localhost:3000/api",
"OOMOL_CONNECT_API_TOKEN": "your-token-here"
}
}
}
}
VSCode
Configure in VSCode settings or use the MCP extension.
Available Tools
High-Level Tools (Recommended)
- list_blocks - List all available blocks (returns latest version by default, optionally all versions)
- list_tasks - List task history
- execute_task - Execute task and wait for completion (most common)
- execute_task_with_files - Execute task with file uploads
- list_packages - List installed packages
- install_package - Install package and wait for completion
Low-Level Tools (Advanced)
- create_task - Create task without waiting (async)
- get_task - Query task status
- stop_task - Stop running task
Usage Examples
List Blocks
{
"tool": "list_blocks",
"arguments": {
"includeAllVersions": false
}
}
Each block automatically includes blockId (format: "package::name") and version fields.
Execute a Task
{
"tool": "execute_task",
"arguments": {
"blockId": "audio-lab::text-to-audio",
"inputValues": {
"text": "Hello",
"voice": "en-US-JennyNeural"
},
"pollIntervalMs": 2000,
"timeoutMs": 300000
}
}
Execute Task with Files
{
"tool": "execute_task_with_files",
"arguments": {
"blockId": "ffmpeg::audio_video_separation",
"inputValues": {
"outputFormat": "mp3"
},
"files": [
{
"path": "/absolute/path/to/video.mp4"
}
]
}
}
Install a Package
{
"tool": "install_package",
"arguments": {
"name": "package-name",
"version": "1.0.0"
}
}
Programmatic Usage
import { OomolConnectMcpServer } from "oomol-connect-mcp-sdk-ts";
const server = new OomolConnectMcpServer({
baseUrl: "http://localhost:3000/api",
authHeader: "Bearer your-token",
name: "my-server",
version: "1.0.0",
});
await server.run();
Development
# Install dependencies
npm install
# Build
npm run build
# Development mode (watch)
npm run dev
# Run locally
npm start
Architecture
This SDK is built on:
@modelcontextprotocol/sdk- Official MCP SDKoomol-connect-sdk- Oomol Connect API client
The SDK provides a zero-logic-duplication wrapper around oomol-connect-sdk, exposing its functionality through the MCP protocol.
License
MIT
Contributing
Contributions are welcome! Please open an issue or submit a pull request.
Related Projects
- oomol-connect-sdk-ts - Core Oomol Connect SDK
- oomol-cloud-mcp-sdk-ts - MCP SDK for Oomol Cloud
Support
For issues and questions:
- GitHub Issues: Create an issue
- Documentation: See examples folder for more usage patterns
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。