Kling Mcp

io.github.runapi-builder由 GitHub 账号 runapi-builder 发布v0.4.1更新于 Oct 29, 2026

RunAPI MCP server for Kling: create tasks, poll status, check pricing.

已验证STDIO仅桌面媒体与设计AI 与机器学习

概览

AI 生成的概览

面向 Kling 模型系列的 RunAPI MCP 服务器:创建视频与数字人生成任务、轮询状态并查询价格。

可以做什么
  • 创建 Kling 视频任务:文生视频、图生视频、视频延长、视频编辑、动作控制与 AI 数字人
  • 查询已有任务的当前状态与最新结果数据
  • 在提交前查询 Kling 模型与端点的 RunAPI 价格
  • 返回任务 ID、状态、输出链接与费用字段,可选择等待任务进入终态
适用场景
当助手需要通过 RunAPI 使用 Kling 模型生成或编辑视频,并且只限定在该模型系列而不加载完整 RunAPI 目录时使用。适合文生视频、图生视频、数字人、动作控制和视频编辑等异步任务流程。
运行要求
通过 npx 从 npm 包以 stdio 方式在本地运行。创建任务和轮询状态需要 RunAPI 账号:使用 login 工具(浏览器 PKCE 登录并写入 RunAPI 配置文件),或在无界面主机上设置 RUNAPI_API_KEY 环境变量。价格查询无需登录。需要访问 RunAPI 的网络连接。
安装前请注意
创建任务会在 RunAPI 账号上产生费用,昂贵或批量请求前应先确认成本。login 工具会打开浏览器并把 API 密钥写入共享的 RunAPI 配置文件,该文件和 RUNAPI_API_KEY 都不应提交到版本库。部分数字人模型 ID 已下线,新请求会返回错误。生成的媒体由第三方服务产出。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Kling Mcp,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

RunAPI Kling MCP Server

Kling API access for AI agents: run video generation operations, poll asynchronous results, and check pricing through one focused MCP server.

Works with Claude Code, Codex, Cursor, Windsurf, VS Code, Roo Code, and any MCP-compatible host.

[npm version] [GitHub repository] [Apache-2.0 license] [MCP Server] [16 models]

Install | Tools | Models | Agent Prompts | Configuration | Links


Why This Package?

@runapi.ai/kling-mcp is a focused Model Context Protocol server for the Kling model line on RunAPI. It gives MCP-compatible assistants direct access to 6 endpoints and 16 model variants without loading the full RunAPI catalog.

Use this per-model server when an agent should stay scoped to Kling. Use @runapi.ai/mcp when one assistant should discover every RunAPI model line.

kling-ai-avatar-v1-pro was retired on 2026-10-14T00:00:00Z. New requests return HTTP 410 before task creation or balance deduction. Existing tasks can still be queried. This model ID is not mapped to another model.

kling-v1-avatar-standard was retired on 2026-10-14T00:00:00Z. New requests return HTTP 410 before task creation or balance deduction. Existing tasks can still be queried. This model ID is not mapped to another model.


Install

Add it to Claude Code:

bash
claude mcp add kling -s user -- npx -y @runapi.ai/kling-mcp

Use project scope when the server should be shared with a repository:

bash
claude mcp add kling -s project -- npx -y @runapi.ai/kling-mcp

Codex, Cursor, Windsurf, VS Code, Roo Code, and other MCP hosts can use the same stdio command:

json
{  "mcpServers": {    "kling": {      "command": "npx",      "args": ["-y", "@runapi.ai/kling-mcp"]    }  }}

check_pricing works before sign-in. For task creation and status polling, ask your assistant to call the login tool. It opens a browser login and saves credentials to ~/.config/runapi/config.json, the same file used by runapi login. Headless and CI hosts can still set RUNAPI_API_KEY before starting the MCP host.

Ready-made examples are in examples/ for Claude, Cursor, Windsurf, VS Code, and Roo Code.


Tools

ToolAuthPurpose
ai_avatarYesCreate a Kling ai avatar task and optionally wait for a terminal status. Returns the task id, status, and output URLs.
edit_videoYesCreate a Kling edit video task and optionally wait for a terminal status. Returns the task id, status, and output URLs.
extend_videoYesCreate a Kling extend video task and optionally wait for a terminal status. Returns the task id, status, and output URLs.
image_to_videoYesCreate a Kling image to video task and optionally wait for a terminal status. Returns the task id, status, and output URLs.
motion_controlYesCreate a Kling motion control task and optionally wait for a terminal status. Returns the task id, status, and output URLs.
text_to_videoYesCreate a Kling text to video task and optionally wait for a terminal status. Returns the task id, status, and output URLs.
get_taskYesFetch the current status and latest payload for an existing task.
check_pricingNoLook up current pricing for a Kling model and endpoint.

Models

Kling covers 16 model variants across 6 endpoints. Each tool accepts the models listed for it:

ToolModels
ai_avatarkling-ai-avatar-pro, kling-ai-avatar-standard
edit_videokling-v3-omni-edit, kling-v3-omni-reference
extend_videokling-v2.5-turbo-image-to-video-pro, kling-v2.5-turbo-text-to-video-pro
image_to_videokling-o1, kling-v2.1-master-image-to-video, kling-v2.1-pro, kling-v2.1-standard, kling-v2.5-turbo-image-to-video-pro, kling-v2.6, kling-v3-omni, kling-v3-turbo-image-to-video
motion_controlkling-3.0, kling-v2.6
text_to_videokling-3.0, kling-o1, kling-v2.1-master-text-to-video, kling-v2.5-turbo-text-to-video-pro, kling-v2.6, kling-v3-omni, kling-v3-omni-reference, kling-v3-turbo-text-to-video

Model availability can change between releases. Use check_pricing or the Kling model page for the current catalog view.


Agent Prompts

Ask your assistant in natural language; it can inspect pricing, create the task, and return the task id plus output URLs.

Create a task

text
Run a Kling ai avatar task with RunAPI.

The assistant can call check_pricing, then ai_avatar, and return the task id, status, and output URLs.

Submit without waiting

text
Create the task but don't wait for it to finish.

The assistant calls the create tool with wait: false and returns the task id. Check on it later with get_task.

Check pricing before creating

text
Check current Kling pricing, then create the task if it matches my request.

The assistant calls check_pricing and can link to the Kling model page for the canonical catalog entry.


Configuration

The server resolves auth in this order:

  1. RUNAPI_API_KEY environment variable, useful for headless and CI hosts
  2. ~/.config/runapi/config.json, created by the MCP login tool or runapi login
  3. No key, which still allows check_pricing

The config file is normally managed by login. A pre-provisioned headless config can use:

json
{  "apiKey": "your_runapi_key"}

Do not commit real API keys.


Links

ResourceURL
Kling model pagehttps://runapi.ai/models/kling
npm package@runapi.ai/kling-mcp
GitHub repositoryrunapi-ai/kling-mcp
RunAPI MCP overviewrunapi.ai/mcp
RunAPI docsrunapi.ai/docs

License

Licensed under the Apache License, Version 2.0.

来源:README.md,提交 a3ed86c

工具

9

login

Authenticate RunAPI by opening a browser PKCE login flow and saving the API key to ~/.config/runapi/config.json.

输入参数force

ai_avatar

Create a Kling task on RunAPI (ai avatar). Returns a task id, status, and output URLs.

输入参数waitmodelprompt必填timeout_mscallback_urlpoll_interval_mssource_audio_url必填source_image_url必填

edit_video

Create a Kling task on RunAPI (edit video). Returns a task id, status, and output URLs.

输入参数waitmodelprompt必填timeout_msaspect_ratiocallback_urlenable_soundsource_task_idduration_secondspoll_interval_mssource_video_urloutput_resolution另有 1 个

extend_video

Create a Kling task on RunAPI (extend video). Returns a task id, status, and output URLs.

输入参数modewaitmodelprompttimeout_mscallback_urlsource_task_id必填poll_interval_ms

image_to_video

Create a Kling task on RunAPI (image to video). Returns a task id, status, and output URLs.

输入参数modewaitmodelprompt必填cfg_scaletimeout_msaspect_ratiocallback_urlenable_soundnegative_promptduration_secondspoll_interval_ms另有 7 个

motion_control

Create a Kling task on RunAPI (motion control). Returns a task id, status, and output URLs.

输入参数waitmodelprompttimeout_mscallback_urlpoll_interval_mssource_image_url必填output_resolution必填reference_video_url必填character_orientation

text_to_video

Create a Kling task on RunAPI (text to video). Returns a task id, status, and output URLs.

输入参数modewaitmodelpromptcfg_scaleshot_typetimeout_msmulti_shotsaspect_ratiocallback_urlenable_soundmulti_prompt另有 11 个

get_task

Fetch the current status and latest result payload for a kling task.

输入参数action必填task_id必填

check_pricing

Look up RunAPI pricing for the kling model line.

输入参数modelaction

版本历史

5
  1. v0.4.1最新Oct 22, 2026
  2. v0.4.0Oct 15, 2026
  3. v0.3.0Sep 30, 2026
  4. v0.2.0Sep 29, 2026
  5. v0.1.14Sep 16, 2026

评分

已满足适用检查项的 90%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分
  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分