RunningHub MCP Server

A FastMCP-based server providing a comprehensive toolset for RunningHub AI.

README.md

RunningHub MCP Server

基于 FastMCP 框架的 MCP 服务,为 RunningHub AI 应用平台提供完整的工具集。

功能概览

  • 节点查询 - 获取 AI 应用的可配置节点列表
  • 文件上传 - 上传图片/音频/视频到 RunningHub 平台
  • 任务提交 - 提交 AI 应用任务并获取 taskId
  • 结果查询 - 查询任务执行状态和输出结果
  • 一键执行 - 提交任务并自动轮询等待完成
  • 任务管理 - 基于本地 JSON 文件的任务持久化管理

安装

# 使用 uv(推荐)
cd runninghub-mcp
uv sync

# 或使用 pip
pip install -e .

配置

通过环境变量配置服务:

环境变量 必需 默认值 说明
RUNNINGHUB_API_KEY - RunningHub API 密钥
RUNNINGHUB_API_HOST www.runninghub.cn API 主机地址
RUNNINGHUB_TASK_STORE_PATH ~/.runninghub/tasks.json 任务持久化文件路径

MCP 客户端配置示例

Cursor / Claude Desktop (mcp.json)

{
  "mcpServers": {
    "runninghub": {
      "command": "uv",
      "args": ["--directory", "/path/to/runninghub-mcp", "run", "runninghub-mcp"],
      "env": {
        "RUNNINGHUB_API_KEY": "your-api-key"
      }
    }
  }
}

Tool 列表

原子工具(API 端点)

Tool 说明 关键参数
get_node_info 获取 AI 应用的可配置节点列表 webapp_id
upload_file 上传文件到 RunningHub file_path
submit_task 提交 AI 应用任务 webapp_id, node_info_list
query_task_outputs 查询任务状态和输出结果 task_id

完整流程工具

Tool 说明
run_task_and_wait 提交任务并轮询等待完成,支持自定义超时和轮询间隔

持久化管理工具

Tool 说明 关键参数
list_tasks 查询本地存储的任务列表 status(可选), limit
get_task_detail 获取指定任务的详细信息 task_id
sync_task_status 从 API 同步任务最新状态到本地 task_id

使用流程

典型的使用流程如下:

  1. 调用 get_node_info 获取 AI 应用的节点列表
  2. 根据需要修改节点参数(如需上传文件,先调用 upload_file
  3. 调用 run_task_and_wait 提交任务并等待结果(或分步使用 submit_task + query_task_outputs
  4. 使用 list_tasks / get_task_detail 查看历史任务

项目结构

src/runninghub_mcp/
  __init__.py     # 包入口
  server.py       # FastMCP 服务实例 + Tool 定义 + 入口
  api.py          # RunningHub HTTP API 封装
  storage.py      # 任务持久化管理(JSON 文件)
  models.py       # 数据模型定义

开发

# 安装开发依赖
uv sync

# 直接运行服务(STDIO 模式)
RUNNINGHUB_API_KEY=your-key uv run runninghub-mcp

# 使用 fastmcp dev 调试
RUNNINGHUB_API_KEY=your-key uv run fastmcp dev src/runninghub_mcp/server.py

Tools 8

get_node_infoGet the list of configurable nodes for an AI application.
upload_fileUpload files to the RunningHub platform.
submit_taskSubmit an AI application task.
query_task_outputsQuery the execution status and output results of a task.
run_task_and_waitSubmit a task and automatically poll until completion.
list_tasksQuery the locally stored task list.
get_task_detailGet detailed information for a specific task.
sync_task_statusSync the latest task status from the API to local storage.

Environment Variables

RUNNINGHUB_API_KEYrequiredRunningHub API key
RUNNINGHUB_API_HOSTAPI host address
RUNNINGHUB_TASK_STORE_PATHPath for task persistence file

Try it

Get the available node information for the AI application with ID 123.
Upload the image at ./my_photo.jpg to RunningHub.
Submit a new task for webapp 123 and wait for it to finish.
List all my recent tasks and check their current status.
Sync the status of task ID 456 to see if it has finished processing.

Frequently Asked Questions

What are the key features of RunningHub MCP Server?

Query AI application node configurations. Upload media files directly to the platform. Submit and manage AI tasks with automated polling. Persistent local storage for task history and status tracking.

What can I use RunningHub MCP Server for?

Automating AI image or video generation workflows via RunningHub. Managing long-running AI tasks without manual status checking. Integrating RunningHub task history into local development environments.

How do I install RunningHub MCP Server?

Install RunningHub MCP Server by running: cd runninghub-mcp && uv sync

What MCP clients work with RunningHub MCP Server?

RunningHub MCP Server works with any MCP-compatible client including Claude Desktop, Claude Code, Cursor, and other editors with MCP support.

Turn this server into reusable context

Keep RunningHub MCP Server docs, env vars, and workflow notes in Conare so your agent carries them across sessions.

Open Conare