Closed rickhardslab closed 11 months ago
Are you running a proxy?
Are you running a proxy?
No proxy, however worth noting this is my first deployment and I've never had it working, it's possible I've failed to set it up properly somehow.
I did set up my google and OpenAI APIs, both paid. The actual googling seems to work, it's only after it decides to summarize the website does it fail to get any results.
Note, ran python -m unittest discover tests
and got:
.........json I will first need to browse the repository (https://github.com/Torantulino/Auto-GPT) and identify any potential bugs that need fixing. I will use the "browse_website" command for this.
{
"command": {
"name": "browse_website",
"args":{
"url": "https://github.com/Torantulino/Auto-GPT"
}
},
"thoughts":
{
"text": "Browsing the repository to identify potential bugs",
"reasoning": "Before fixing bugs, I need to identify what needs fixing. I will use the 'browse_website' command to analyze the repository.",
"plan": "- Analyze the repository for potential bugs and areas of improvement",
"criticism": "I need to ensure I am thorough and pay attention to detail while browsing the repository.",
"speak": "I am browsing the repository to identify potential bugs."
}
}
json loads error Expecting value: line 1 column 1 (char 0)
.json BEGIN: "name": "John" - "age": 30 - "city": "New York" :END
json loads error Expecting value: line 1 column 1 (char 0)
.json BEGIN: "name": "John" - "age": 30 - "city": "New York" :END
json loads error Expecting value: line 1 column 1 (char 0)
.json {"name": "John", "age": 30, "city": "New York",}
json loads error Expecting property name enclosed in double quotes: line 1 column 48 (char 47)
.......... ←[0m Warning: model not found. Using cl100k_base encoding.
.....
----------------------------------------------------------------------
Ran 27 tests in 0.216s
OK
The version of AutoGPT I configured was the stable branch downloaded on 4/16/23 @ 8:30pm EST. If someone can point me to a version lookup I can provide that instead.
This issue has automatically been marked as stale because it has not had any activity in the last 50 days. You can unstale it by commenting or removing the label. Otherwise, this issue will be closed in 10 days.
This issue was closed automatically because it has been stale for 10 days with no activity.
⚠️ Search for existing issues first ⚠️
GPT-3 or GPT-4
Steps to reproduce 🕹
When choosing a prompt which will involve AutoGPT to browse the internet, it fails to return the results with an error.
Every test I've performed resulted in AutoGPT to choose the browse_website command.
Current behavior 😯
AutoGPT will choose to use the browse_website command, but fails to return results:
Note the latter SYSTEM: output stating the failure. If allowed to execute, AutoGPT will choose to google again, attempt to open the most relevant results, and fail with the same error and code 100% of the time.
Expected behavior 🤔
When AutoGPT chooses to use browse_website, it should return usable results without error.
Your prompt 📝
last_run_ai_settings.yaml does not exist within my AutoGPT directory even after running the above tests. Prompt details included in the 'reproduce' section.