[BUG]assitant promt cause llm produce fake content

[J-lena]

Description

We do not want the model to generate an Observation during tool calls，beause when we use flow streaming ,we can accept this unused content. We have already customized the tools prompt, but the assistant prompt is automatically added, which causes subsequent tool calls to fabricate the tool’s output

Steps to Reproduce

when use tools

Expected behavior

when usel tools,Observation shoued not generated by llm

Screenshots/Code snippets

Operating System

Ubuntu 20.04

Python Version

3.10

crewAI Version

latest

crewAI Tools Version

xx

Virtual Environment

Venv

Evidence

xx

Possible Solution

xx

Additional context

xx