So I have a interesting setup. i big Unraid Server with a very small 5GB Nvidia Quadro P2200. My Ollama VM is 100GB RAM and has the Nvidia Quadro P2200 passed through to yet. I then have a Agent-Zero Docker container latest as of yesterday. The Models are Chat = qwen2.5:3b Utility = qwen2.5:1.5b When as it to create an create an ebook I get the following error. ValueError: Tool request must have a tool_name (type string) field Traceback (most recent call last): Traceback (most recent call last): File "/a0/helpers/extension.py", line 176, in _run_async data["result"] = await data["result"] ^^^^^^^^^^^^^^^^^^^^ File "/a0/agent.py", line 572, in handle_exception raise exception # exception handling is done by extensions ^^^^^^^^^^^^^^^ File "/a0/agent.py", line 485, in monologue tools_result = await self.process_tools(agent_response) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/a0/helpers/extension.py", line 183, in _run_async result = _process_result(data) ^^^^^^^^^^^^^^^^^^^^^ File "/a0/helpers/extension.py", line 143, in _process_result raise exc File "/a0/helpers/extension.py", line 176, in _run_async data["result"] = await data["result"] ^^^^^^^^^^^^^^^^^^^^ File "/a0/agent.py", line 854, in process_tools await self.validate_tool_request(tool_request) File "/a0/helpers/extension.py", line 183, in _run_async result = _process_result(data) ^^^^^^^^^^^^^^^^^^^^^ File "/a0/helpers/extension.py", line 143, in _process_result raise exc File "/a0/helpers/extension.py", line 176, in _run_async data["result"] = await data["result"] ^^^^^^^^^^^^^^^^^^^^ File "/a0/agent.py", line 956, in validate_tool_request raise ValueError("Tool request must have a tool_name (type string) field")

Did any of you have any luck in creating anything in blender. Using Agent Zero MCP? a cube or something.