LumaMCP

Instructions

This allows you to continue a previously generated video, adding more motion
and content after the original video ends.

Use this when:
- A generated video is too short and you want to add more
- You want to continue the story or motion from a previous video
- You're building a longer video piece by piece

Returns:
    Task ID and the extended video information.

Implementation Reference

• tools/video_tools.py:155-199 (handler)

  Main handler function for luma_extend_video tool. Extends an existing video by its ID with additional content using the Luma API. Takes video_id, prompt, and optional end_image_url as parameters.

  @mcp.tool()
  async def luma_extend_video(
      video_id: Annotated[
          str,
          Field(
              description="ID of the video to extend. This is the 'video_id' field from a previous generation result."
          ),
      ],
      prompt: Annotated[
          str,
          Field(
              description="Description of what should happen in the extended portion of the video. Describe the continuation of motion and new content."
          ),
      ],
      end_image_url: Annotated[
          str,
          Field(
              description="Optional URL of an image to use as the final frame of the extended video."
          ),
      ] = "",
  ) -> str:
      """Extend an existing video with additional content.
  
      This allows you to continue a previously generated video, adding more motion
      and content after the original video ends.
  
      Use this when:
      - A generated video is too short and you want to add more
      - You want to continue the story or motion from a previous video
      - You're building a longer video piece by piece
  
      Returns:
          Task ID and the extended video information.
      """
      payload = {
          "action": "extend",
          "video_id": video_id,
          "prompt": prompt,
      }
  
      if end_image_url:
          payload["end_image_url"] = end_image_url
  
      result = await client.generate_video(**payload)
      return format_video_result(result)

• tools/video_tools.py:155-175 (schema)

  Input schema for luma_extend_video defined via Pydantic Annotated Field types. Defines video_id (required str), prompt (required str), and end_image_url (optional str with default='').

  @mcp.tool()
  async def luma_extend_video(
      video_id: Annotated[
          str,
          Field(
              description="ID of the video to extend. This is the 'video_id' field from a previous generation result."
          ),
      ],
      prompt: Annotated[
          str,
          Field(
              description="Description of what should happen in the extended portion of the video. Describe the continuation of motion and new content."
          ),
      ],
      end_image_url: Annotated[
          str,
          Field(
              description="Optional URL of an image to use as the final frame of the extended video."
          ),
      ] = "",
  ) -> str:

• core/server.py:42-49 (registration)

  MCP server initialization where the FastMCP instance is created. The @mcp.tool() decorator used in video_tools.py registers tools with this mcp instance.

  # Initialize FastMCP server
  mcp = FastMCP(
      settings.server_name,
      icons=[Icon(src="https://cdn.acedata.cloud/ahjfwi.png")],
      **mcp_kwargs,
  )
  
  logger.info(f"Initialized MCP server: {settings.server_name}")

• tools/__init__.py:1-11 (registration)

  Tools module initialization that imports video_tools (and others), causing the @mcp.tool() decorators to execute and register all tools with the MCP server.

  """Tools module for MCP Luma server."""
  
  # Import all tools to register them with the MCP server
  from tools import info_tools, task_tools, video_tools
  
  __all__ = [
      "video_tools",
      "task_tools",
      "info_tools",
  ]

• main.py:112-116 (registration)

  Main entry point where the tools module is imported, triggering the registration of luma_extend_video and other tools via their @mcp.tool() decorators.

  # Import tools and prompts to register them
  safe_print("  Loading tools and prompts...")
  import prompts  # noqa: F401, I001
  import tools  # noqa: F401