mcp-kling
🎬 The FIRST MCP server for Kling AI video generation! Generate stunning AI videos directly from Claude.
claude mcp add --transport stdio 199-mcp-mcp-kling npx -y mcp-kling@latest \ --env KLING_ACCESS_KEY="YOUR_ACCESS_KEY_HERE" \ --env KLING_SECRET_KEY="YOUR_SECRET_KEY_HERE"
How to use
MCP Kling is an all-in-one MCP server that exposes Kling AI capabilities through Claude Desktop. It provides access to Kling's video, image, effects, lip-syncing, virtual try-on, and account management features via a single MCP interface. With this server you can generate videos and images, apply professional effects, perform lip-sync with TTS, extend videos, and manage your Kling credits and resource packages. The server handles token generation and communicates with Kling’s APIs to deliver downloads automatically for finished assets. Use the included 13 tools to orchestrate end-to-end creative workflows from text prompts to final media assets.
To use it, configure Claude Desktop to point at the mcp-kling MCP server (via the npx-based installation). After starting Claude Desktop, you can issue natural language prompts that map to specific tools like generate_video, create_lipsync, generate_image, virtual_try_on, and get_account_balance. The system will execute the requested task, fetch the results, and automatically download media to your local downloads directory when available.
How to install
Prerequisites
- Node.js and npm (or npx) installed on your machine
- Access to Kling API keys (Access Key and Secret Key)
- Claude Desktop client installed to configure MCP servers
Step-by-step installation
- Install the MCP Kling server via npx (no global installation required):
npx -y mcp-kling@latest
- When prompted, provide your Kling credentials or set them via environment variables in your Claude Desktop config (see config example below).
- Add the MCP server to Claude Desktop:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%\Claude\claude_desktop_config.json
- Restart Claude Desktop and begin issuing prompts to access Kling tools.
Notes
- You only need to run the npx command once to install and run the MCP Kling server locally.
- Ensure your Kling API keys are kept secure and not hard-coded in shared configs.
- The MCP server will automatically download generated content to your configured downloads location.
Additional notes
Tips and common issues:
- Environment variables: KLING_ACCESS_KEY and KLING_SECRET_KEY must be set for authentication with Kling.
- If you see token or authentication errors, re-check your Kling keys and ensure they have proper permissions.
- The MCP supports automatic downloads; verify your downloads path is writable by the CLI/Claude Desktop.
- While using generate_video or generate_image, consider model availability (v1.0, v1.5, v1.6 for Kling) and allowed aspect ratios to avoid failures.
- For long tasks, you can use check_video_status or check_image_status to monitor progress and auto-download when ready.
Related MCP Servers
MiniMax
Official MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
boilerplate
TypeScript Model Context Protocol (MCP) server boilerplate providing IP lookup tools/resources. Includes CLI support and extensible structure for connecting AI systems (LLMs) to external data sources like ip-api.com. Ideal template for creating new MCP integrations via Node.js.
claude-code
MCP Server connects with claude code local command.
PixVerse
Official PixVerse Model Context Protocol (MCP) server that enables interaction with powerful AI video generation APIs.
storybook
MCP server for Storybook - provides AI assistants access to components, stories, properties and screenshots. Built with TypeScript and Model Context Protocol SDK.
miro
MCP server for controlling Miro whiteboards with AI assistants