Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@RunningHub MCP Server Run the AI task for webapp ID 1024 and wait for the results."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
基于 FastMCP 框架的 MCP 服务，为 RunningHub AI 应用平台提供完整的工具集。
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
Related MCP server: MCP Server for Replicate
# 使用 uv（推荐）
cd runninghub-mcp
uv sync
# 或使用 pip
pip install -e .通过环境变量配置服务：
环境变量 | 必需 | 默认值 | 说明 |
| 是 | - | RunningHub API 密钥 |
| 否 |
| API 主机地址 |
| 否 |
| 任务持久化文件路径 |
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 | 说明 | 关键参数 |
| 获取 AI 应用的可配置节点列表 |
|
| 上传文件到 RunningHub |
|
| 提交 AI 应用任务 |
|
| 查询任务状态和输出结果 |
|
Tool | 说明 |
| 提交任务并轮询等待完成，支持自定义超时和轮询间隔 |
Tool | 说明 | 关键参数 |
| 查询本地存储的任务列表 |
|
| 获取指定任务的详细信息 |
|
| 从 API 同步任务最新状态到本地 |
|
典型的使用流程如下：
调用 get_node_info 获取 AI 应用的节点列表
根据需要修改节点参数（如需上传文件，先调用 upload_file）
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）
使用 list_tasks / get_task_detail 查看历史任务
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# 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.pyGet the configurable node list of an AI application. Returns all modifiable node info (nodeId, fieldName, fieldType, fieldValue).
| Name | Required | Description | Default |
|---|---|---|---|
| webapp_id | Yes | The webapp ID of the AI application |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden. It indicates a read operation returning modifiable node info but lacks details on idempotence, auth requirements, rate limits, or potential side effects.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, no fluff. The first sentence states the primary action, and the second enumerates the return fields. Information is front-loaded and efficiently structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the simple retrieval nature, the description covers the essential purpose and return fields. An output schema exists but is not shown; however, the description lists the fields. No discussion of errors, pagination, or examples, but adequate for the tool's simplicity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with a well-described parameter. The description adds the context that the list pertains to an AI app but does not provide additional parameter constraints or formatting beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it retrieves a list of configurable nodes for an AI application, specifying the exact fields returned. This verb-resource combination is distinct from sibling tools, which focus on tasks.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies this tool is for retrieving node configurations of an AI app but does not explicitly state when to use it over alternatives or provide conditions for avoidance. No guidance on prerequisites or context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get detailed information of a specific task from local persistent storage.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes | Task ID |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided. Description does not disclose important behavioral traits like whether it's read-only, whether it may throw errors for invalid IDs, or performance characteristics. Only states it gets info from local storage.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single sentence, no fluff. All essential information is front-loaded, efficient use of words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the existence of an output schema (not detailed here), the description adequately covers purpose. Could mention what 'detailed information' entails, but overall sufficient for a simple retrieval tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% for the single parameter 'task_id'. The description 'Task ID' is clear but adds no extra meaning beyond the schema. Baseline of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Get detailed information'), the resource ('a specific task'), and the source ('from local persistent storage'). It distinguishes from sibling tools like list_tasks (which lists tasks) and get_node_info (node-level info).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use or avoid this tool vs alternatives. Doesn't mention prerequisites, limitations, or context like needing a valid task_id.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List tasks from local persistent storage. Supports filtering by status (pending/queued/running/completed/failed). Results are sorted by last update time in descending order.
| Name | Required | Description | Default |
|---|---|---|---|
| status | No | Filter by status: pending, queued, running, completed, failed (optional) | |
| limit | No | Max number of results |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Without annotations, the description carries full burden. It discloses data source, filtering, and sorting, which are key. However, it lacks details on whether the operation has side effects (likely none), authentication needs, or pagination behavior beyond the limit parameter.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences with no redundancy. The most important information (action, source, filter, sort) is front-loaded and clear.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the presence of an output schema (which covers return values) and the relatively simple function, the description provides all necessary context: what it does, what filtering is available, and how results are ordered. No critical gaps remain.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3. The description restates filtering by status and adds sorting context, but does not substantially augment the schema definitions for the two parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's action (list tasks), source (local persistent storage), filtering by status, and sorting order. It effectively distinguishes from sibling tools like get_task_detail or query_task_outputs which serve different specific purposes.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for browsing or filtering tasks, but does not explicitly state when to use this tool over alternatives. However, given sibling names, the context is clear enough for an agent to infer.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Query the execution status and output of a task. Status codes: 0=completed, 804=running, 813=queued, 805=failed. Returns output file URLs when completed.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes | Task ID returned by submit_task |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so description bears full burden. It reveals status codes and output URLs, but omits potential aspects like error response structure, rate limits, or authentication requirements.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, no fluff, front-loaded with purpose. Efficient and to the point.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool is simple with one parameter and output schema exists. Description covers status codes and output URLs. Could mention failure details, but overall adequate.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% for task_id, so description adds no further parameter meaning beyond what schema states. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states tool queries execution status and output of a task, lists status codes, and mentions output file URLs on completion. Distinguishes from siblings like submit_task and run_task_and_wait.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Description implies use after task submission, but no explicit when-to-use or when-not-to-use guidance. Does not compare with sibling run_task_and_wait which blocks.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Submit an AI application task and wait for completion (auto-polling). Combines submit_task + query_task_outputs into a full workflow. Default max wait is 10 minutes with 5-second polling interval. Task status is automatically synced to persistent storage.
| Name | Required | Description | Default |
|---|---|---|---|
| webapp_id | Yes | The webapp ID of the AI application | |
| node_info_list | Yes | Node parameter list, each containing nodeId, fieldName, fieldType, fieldValue | |
| timeout | No | Max wait time in seconds | |
| poll_interval | No | Polling interval in seconds |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description discloses important behaviors: auto-polling, default timeout (10 minutes), polling interval (5 seconds), and automatic syncing to persistent storage. It does not cover failure behavior or error handling, but the provided details are substantial.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is three sentences long, efficiently conveying purpose, behavior, and defaults. It is front-loaded with the main action. Minor improvements could be made by combining related information, but it is already concise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity and the existence of an output schema, the description covers the core workflow, polling behavior, and persistence. It lacks details on timeout handling or task failure, but overall it is sufficiently complete for an agent to use effectively.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the description does not need to add much. It repeats default values for timeout and poll_interval already in the schema, adding minimal extra semantics. Baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's action: 'Submit an AI application task and wait for completion (auto-polling)'. It explains that it 'Combines submit_task + query_task_outputs into a full workflow', which distinguishes it from sibling tools like submit_task and query_task_outputs.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies when to use this tool (as a full workflow combining submission and polling) but does not explicitly state when not to use it or mention alternatives. The context is clear enough for an AI agent to infer appropriate usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Submit an AI application task. Requires webapp_id and node_info_list. Returns taskId for subsequent status queries. The task is automatically saved to local persistent storage.
| Name | Required | Description | Default |
|---|---|---|---|
| webapp_id | Yes | The webapp ID of the AI application | |
| node_info_list | Yes | Node parameter list, each containing nodeId, fieldName, fieldType, fieldValue |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden. It discloses that the task is automatically saved to local persistent storage, which is helpful, but omits other behavioral traits like asynchronicity, authentication needs, or side effects beyond persistence.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with three sentences. The first sentence immediately states the purpose, followed by key requirements and behavior. No extraneous information is present.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity, the description covers the essential aspects: purpose, required parameters, return value, and persistence. The existence of an output schema reduces the need to detail return values further. However, it could mention asynchronicity or rate limits for full completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 100% coverage with descriptions for both parameters. The description merely restates the requirement of these parameters without adding meaning beyond the schema. The mention of 'node_info_list' as a list of objects is already in the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states 'Submit an AI application task,' using a specific verb and resource. It distinguishes itself from sibling tools like get_node_info and list_tasks by being the only creation tool, and mentions required parameters and return value.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description specifies required parameters (webapp_id, node_info_list) and explains the return value (taskId) for subsequent status queries. While it implicitly contrasts with siblings (e.g., sync_task_status for status), it lacks explicit when-to-use guidance or alternative mentions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Sync task status from RunningHub API to local persistent storage. Useful for refreshing status of previously timed-out tasks.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes | Task ID to sync |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must fully disclose behavioral traits. It indicates a write operation ('to local persistent storage') but does not mention idempotency, side effects, or authentication requirements. Minimal disclosure beyond basic function.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, efficient and front-loaded with action. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool is simple with one parameter and has an output schema, so description need not detail return values. Covers purpose and use case adequately. Could optionally mention if the operation is synchronous, but not required.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% (task_id described). Description adds context about use case but does not enhance parameter meaning beyond schema. Baseline 3 is appropriate as schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states the tool syncs task status from API to local storage, specifically for refreshing timed-out tasks. Verb 'sync' and resource 'task status' are specific, and the use case distinguishes from sibling tools that retrieve or submit tasks.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Description mentions 'useful for refreshing status of previously timed-out tasks,' which implies when to use, but does not explicitly state when not to use or provide alternatives. No comparison with sibling tools like get_task_detail or list_tasks.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Upload a local file to RunningHub platform. Supports IMAGE, AUDIO, VIDEO types. Returns the uploaded fileName for use in node fieldValue.
| Name | Required | Description | Default |
|---|---|---|---|
| file_path | Yes | Absolute path of the local file |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so description carries full burden. Only states basic upload function and supported types. Lacks disclosure on overwrite behavior, size limits, permissions, or failure handling.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences that front-load purpose and key details. No unnecessary words. Efficiently structured.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple one-parameter upload tool, the description covers purpose, supported types, and return value. Minor gaps in behavioral details (e.g., error handling) but overall adequate given low complexity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema has one parameter with description. Description adds value by specifying supported file types (IMAGE, AUDIO, VIDEO), which are not in schema enums. This provides extra context beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Clearly states verb 'upload', resource 'local file to RunningHub platform', and supported types (IMAGE, AUDIO, VIDEO). Distinguishes from sibling tools (focused on tasks and nodes).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance on when to use this tool versus alternatives. It does not mention prerequisites or context for usage beyond the basic action.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0get_node_infoget_task_detaillist_tasksquery_task_outputsrun_task_and_waitsubmit_tasksync_task_statusupload_fileScored across 8 tools
There is overlap between submit_task and run_task_and_wait (both submit tasks), and between query_task_outputs and run_task_and_wait (both check status/output). However, descriptions clarify the combination pattern in run_task_and_wait, so ambiguity is limited but still present.
Most tools follow a consistent snake_case verb_noun pattern (e.g., get_node_info, submit_task). The exception is run_task_and_wait, which deviates with 'and_wait', and 'list_tasks' vs 'query_task_outputs' use different verbs for similar operations.
8 tools is well-scoped for an AI task runner, covering submission, synchronization, file upload, and status queries without being excessive or minimal.
Core workflow (submit/run, query output, sync status, upload files) is covered. Missing tools for listing available applications or canceling tasks, but these are not critical for the primary use case.
Remote MCP server for AIStoryHub: stories, chapters, story bible, Voiceprints, AI generation.
Hosted MCP server for AI-driven data ops. Create apps, manage schemas, and CRUD structured data.
MCP server for QPost — lets AI agents publish video and image posts to YouTube, TikTok, Instagram.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
