Any idea how to fix this one ? I’m using a custom tool built using BaseTool
Can you share your tools code and the crew agent that is running it?
Seeing the same error here when running FileReadTool so my agent can read a csv file. It was working last week and when I was running my code again today, it is throwing this error:
I encountered an error while trying to use the tool. This was the error: StructuredTool._run() missing 1 required keyword-only argument: ‘config’.
Tool Ask question to co-worker accepts these inputs: Ask a specific question to one of the following co-workers: [My Agent]
Did you update or change anything? Are you running which OS. Windows can have some weird file path issues vs OSX or linux. Also can you share a bit more info so we can figure this out please?
Sounds like an llm just making an error. Did you run it with gpt 4 ?
I am running my code on a Mac, so Mac OS. And using gpt 4-turbo.
There is nothing fancying on my code, just trying to have the agent to read a csv file on my computer.
Ok so probably not the llm. Need to see the tool code and the verbose=true messages statements. Is it setting up the right action input, etc.
Also is the tool setup in Agent or Task or both. Try all 3 different approaches.
So the tool code is just an import from crewai_tools to get FileReadTool. Verbose = True and tools are setup in the Agent. Thought it was not possible at the task level…
Heard something in other forums that Langchain changed something in their StructuredTool._run() which might got me thinking that it might be the caused of this error?
You can setup the tool in Task. Try all 3. What are the messages for action input, etc. is it setting up correctly to use the tool parameter wise? Where Is it failing after it says its going to use the tool? Where exactly is the error appearing?