search

gptscript-ai / gptscript

Build AI assistants that interact with your systems
https://gptscript.ai
Apache License 2.0
2.95k stars 260 forks source link

UI - Attempt to prompt for entering provider credentials when launching UI fails #513

Open sangee2004 opened 2 months ago

sangee2004 commented 2 months ago

gptscript version v0.8.2+1b5e068f

Steps to reproduce the problem:

  1. Launch UI using a model from anthropic-provider
%gptscript --default-model 'claude-3-opus-20240229  from github.com/gptscript-ai/anthropic-provider'  --ui            
14:22:49 WARNING: Changing the default model can have unknown behavior for existing tools. Use the model field per tool instead.
14:22:49 started  [main]
14:22:49 started  [provider: https://raw.githubusercontent.com/gptscript-ai/anthropic-provider/3b9b9365d469c4c702291f1764537ab5c226c2e0/tool.gpt:Anthropic Claude3 Model Provider]
14:22:49 started  [credential: github.com/gptscript-ai/claude3-anthropic-provider/credential]
14:22:49 sent     [credential: github.com/gptscript-ai/claude3-anthropic-provider/credential]
? Please enter your Anthropic API token. ^[[44;219R^[[44;42R2024/06/18 14:22:50 getting prompt response invalid status code [500], expected 200
Failed to run sys.prompt.
14:22:50 failed to run tool [claude3-anthropic-credential] cmd [/usr/bin/env python3 /Users/sangeethahariharan/Library/Caches/gptscript/repos/3b9b9365d469c4c702291f1764537ab5c226c2e0/credential/tool.gpt/python3.12/credential/credential.py]: exit status 1
2024/06/18 14:22:50 no prompt server found, can not continue
sangeethahariharan@Sangeethas-MBP gptscript % 4;219R4;42R

  1. Launch UI .

    sangeethahariharan@Sangeethas-MBP gptscript % gptscript --ui        
14:24:54 started  [main]
2024/06/18 14:24:54 no prompt server found, can not continue

  2. Launch UI using a model from local LLM 

gptscript --disable-cache --default-model 'Llama-3-8b-function-calling-alpha-v1.gguf from http://localhost:1234/v1' --ui examples/bobchat.gpt
14:27:02 WARNING: Changing the default model can have unknown behavior for existing tools. Use the model field per tool instead.
14:27:03 started  [main] [input=--file=bobchat.gpt]
2024/06/18 14:27:03 no prompt server found, can not continue
dbpolito commented 1 month ago

i'm facing a similar issue:

```gptscript --ui
11:56:01 started  [main]
11:56:01 started  [context: github.com/gptscript-ai/context/os]
11:56:01 sent     [context: github.com/gptscript-ai/context/os]
11:56:01 ended    [context: github.com/gptscript-ai/context/os] [output=The local operating systems is uname_result(system='Darwin', node='MacBook-Pro-de-Daniel.local', rel...]
2024/07/30 11:56:01 no prompt server found, can not continue