Question Validation

• I have searched both the documentation and discord for an answer.

Question

I couldn't find this in the docs. In the latest version of the docs, the multi-agent report generation using agents as tools example (https://developers.llamaindex.ai/python/examples/agent/agents_as_tools/) uses plain old Python functions instead of wrapping them in a FunctionTool. What is the use-case for using a FunctionTool versus a Python function, and are there any gotchas with using one versus the other?