mcp-luma-dream-machine

Instructions

Implementation Reference

• src/luma_ai_mcp_server/server.py:306-330 (handler)

  The main handler function that executes the get_generation tool logic: fetches generation status from Luma API using the provided generation_id and formats the response.

  async def get_generation(parameters: dict) -> str:
      """Get the status of a generation."""
      try:
          generation_id = parameters.get("generation_id")
          if not generation_id:
              raise ValueError("generation_id parameter is required")
  
          result = await _make_luma_request("GET", f"/generations/{generation_id}")
  
          if not isinstance(result, dict):
              raise ValueError("Invalid response from API")
  
          output = [f"Generation ID: {result['id']}", f"State: {result['state']}"]
  
          if result.get("failure_reason"):
              output.append(f"Reason: {result['failure_reason']}")
  
          if result.get("assets", {}).get("video"):
              output.append(f"Video URL: {result['assets']['video']}")
  
          return "\n".join(output)
      except Exception as e:
          logger.error(f"Error in get_generation: {str(e)}", exc_info=True)
          return f"Error getting generation {generation_id}: {str(e)}"

• src/luma_ai_mcp_server/server.py:136-137 (schema)

  Pydantic input schema for the get_generation tool, defining the required 'generation_id' field.

  class GetGenerationInput(BaseModel):
      generation_id: str

• src/luma_ai_mcp_server/server.py:508-512 (registration)

  Tool registration in the MCP server's list_tools() method, defining name, description, and input schema for get_generation.

  Tool(
      name=LumaTools.GET_GENERATION,
      description="Gets the status of a generation",
      inputSchema=GetGenerationInput.model_json_schema(),
  ),

• src/luma_ai_mcp_server/server.py:563-565 (registration)

  Dispatch/handling of get_generation tool call in the MCP server's call_tool() method.

  case LumaTools.GET_GENERATION:
      result = await get_generation(arguments)
      return [TextContent(type="text", text=result)]