Guess
Guess
Are you running in the Windows Terminal or in a tool like MinGW that simulates a Linux runtime environment?
I tried but couldn't reproduce your issue. My environment setup: 1. windows 11 2. python 3.9 3. cygwin64 4. virtualenv 5. MetaGPT main branch
Your API calls are too fast, and have been rate-limited: ```text anthropic.RateLimitError: Error code: 429 - {'type': 'error', 'error': {'type': 'rate_limit_error', 'message': 'Number of requests has exceeded your rate limit...
`Engineer`并未订阅`RunCode`消息，因此即使消息发给了`Engineer`，在`_observe`阶段也会被忽略掉。 建议参考Product Manager给Engineer转bug单的流程，将需要Engineer重写的代码作为bug单转给Engineer。参见`metagpt/actions/write_prd.py`的`_handle_bugfix`： ```python async def _handle_bugfix(self, req: Document) -> Message: # ... bugfix logic ... await self.repo.docs.save(filename=BUGFIX_FILENAME, content=req.content) await self.repo.docs.save(filename=REQUIREMENT_FILENAME, content="") return AIMessage( content=f"A new issue is received: {BUGFIX_FILENAME}",...
The reason for the issue is that LLM did not return the data in the required format: the "Programming Language" field is missing. This problem can only be resolved by...
The response from the large model is missing the field corresponding to "Required Python packages".
`generate_repo` is not an async function, so it shouldn't appear within `asyncio.run()`. ```python #!/usr/bin/env python # -*- coding: utf-8 -*- import asyncio from pathlib import Path import typer from metagpt.const...
How are you using generate_repo? Could you provide the complete code snippet?
You are not on python 3.9, but python 3.12: ```text File "/home/wuy/software/envs/metagpt/lib/python3.12/site-packages/... ``` python 3.9 or 3.10 is recommended. There are too many issues with Python 3.12 regarding dependencies.