AutoGPT: Can not access *.py files for execution

Duplicates

I have searched the existing issues

Steps to reproduce 🕹

… EXECUTE_LOCAL_COMMANDS=True … python3 scripts/main.py --continuous --gpt3only y

Current behavior 😯

NEXT ACTION:  COMMAND = execute_python_file ARGUMENTS = {'file': 'evaluate_net_worth_strategies.py'}
Executing file 'evaluate_net_worth_strategies.py' in workspace 'auto_gpt_workspace'
SYSTEM:  Command execute_python_file returned: Error: Error while fetching server API version: ('Connection aborted.', FileNotFoundError(2, 'No such file or directory'))

But 'evaluate_net_worth_strategies.py exists! in Auto-GPT/auto_gpt_workspace ! I swear

Expected behavior 🤔

*.py file execution

Your prompt 📝

# Paste your prompt here

About this issue

Original URL
State: closed
Created a year ago
Comments: 15 (5 by maintainers)

Most upvoted comments

The issue is that someone rewrote the execute_python_file command to assume it was in a docker container. An app expecting to be in a docker container that isn’t throws a file not found error. Was going nuts from this. Just now had ChatGPT rip out the docker crap for me. Here’s my execute_code.py file: execute_code - execute_python_file fixed.py.txt

Stunspot on Apr 17, 2023