ComfyUI MCP Server

Instructions

    Args:
        workflow: Existing workflow dict
        node_id: Unique identifier for this node
        node_type: Node class name (use list_nodes() to find)
        inputs: Input values. For connections use ["source_node_id", output_index].

    Examples:
        # Simple value input
        add_node(wf, "1", "StringInput_fal", {"text": "a cat"})

        # Connection to another node
        add_node(wf, "2", "CLIPTextEncode", {
            "text": "prompt",
            "clip": ["1", 0]  # Connect to node "1" output 0
        })

    Returns the modified workflow dict.
    

Implementation Reference

• src/comfy_mcp_server/tools/workflow.py:353-386 (handler)

  The primary handler function for the 'add_node' MCP tool. It adds a new node to the provided workflow dictionary by setting workflow[node_id] = {"class_type": node_type, "inputs": inputs}. Includes input schema via Pydantic Field descriptions and registration via @mcp.tool() decorator.

  @mcp.tool()
  def add_node(
      workflow: dict = Field(description="Workflow dict to modify"),
      node_id: str = Field(description="Unique node ID (e.g., '1', 'prompt')"),
      node_type: str = Field(description="Node class name"),
      inputs: dict = Field(description="Node inputs"),
      ctx: Context = None,
  ) -> dict:
      """Add a node to a workflow.
  
      Args:
          workflow: Existing workflow dict
          node_id: Unique identifier for this node
          node_type: Node class name (use list_nodes() to find)
          inputs: Input values. For connections use ["source_node_id", output_index].
  
      Examples:
          # Simple value input
          add_node(wf, "1", "StringInput_fal", {"text": "a cat"})
  
          # Connection to another node
          add_node(wf, "2", "CLIPTextEncode", {
              "text": "prompt",
              "clip": ["1", 0]  # Connect to node "1" output 0
          })
  
      Returns the modified workflow dict.
      """
      if ctx:
          ctx.info(f"Adding node {node_id}: {node_type}")
  
      workflow[node_id] = {"class_type": node_type, "inputs": inputs}
      return workflow

• src/comfy_mcp_server/tools/__init__.py:23-28 (registration)

  The register_all_tools function which indirectly registers the add_node tool by calling register_workflow_tools(mcp), where the tool is defined.

  def register_all_tools(mcp):
      """Register all tools with the MCP server."""
      register_system_tools(mcp)
      register_discovery_tools(mcp)
      register_workflow_tools(mcp)
      register_execution_tools(mcp)

• src/comfy_mcp_server/models.py:123-127 (helper)

  A helper method on the Workflow Pydantic model class that adds a node to the workflow's nodes dict, similar logic to the tool handler but typed.

  def add_node(self, node_id: str, class_type: str, inputs: dict) -> "Workflow":
      """Add a node to the workflow."""
      self.nodes[node_id] = WorkflowNode(class_type=class_type, inputs=inputs)
      return self