Skip to content

Credentials - When there are parallel tool calls made , there are texts relating to tool calls that gets displayed after the prompt (when waiting for user input) which is confusing. #256

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
sangee2004 opened this issue Apr 16, 2024 · 1 comment
Assignees
Labels
bug Something isn't working

Comments

@sangee2004
Copy link
Contributor

gptscript version v0.0.0-dev-bc620f0e-dirty

Steps to reproduce the problem:

  1. Run examples/story-book/story-book.gpt which makes multiple parallel tool calls to image-generation tool.

  2. Notice that there are texts relating to tool calls that gets displayed after the prompt "?token" which is confusing.

gptscript --disable-cache examples/story-book/story-book.gpt --story "My trip to Japan" --pages 4

13:04:29 sent     [story-illustrator(11)]
13:04:29 sent     [story-illustrator(9)]
13:04:29 sent     [story-illustrator(8)]
         content  [8] content | Waiting for model response...
13:04:29 sent     [story-illustrator(10)]
         content  [8] content | tool call image-generation -> {"prompt":"In Kyoto, Jamie wanders through ancient streets li ...
13:04:32 started  [credential: github.com/gptscript-ai/image-generation/credential]
13:04:32 sent     [credential: github.com/gptscript-ai/image-generation/credential]
Please enter your OpenAI API token.
? token          content  [8] content | tool call image-generation -> {"prompt":"In Kyoto, Jamie wanders through ancient streets li ...
         content  [8] content | tool call image-generation -> {"prompt":"In Kyoto, Jamie wanders through ancient streets li ...         content  [11] content | tool call image-generation -> {"prompt": "Jamie in Tokyo, surrounded by skyscrapers and arc ...
         content  [10] content | tool call image-generation -> {"prompt":"Jamie in the heart of the countryside, climbing mi ...
         content  [9] content | tool call image-generation -> {"prompt":"Jamie at the Sakura festival in a village, surroun ...
         content  [9] content | tool call image-generation -> {"prompt":"Jamie at the Sakura festival in a village, surroun ...
***************************************************
13:06:47 ended    [credential: github.com/gptscript-ai/image-generation/credential]
13:06:47 started  [story-illustrator(11)->image-generation(13)] [input={"prompt": "Jamie in Tokyo, surrounded by skyscrapers and arcade games. In a pointilism cartoon children's book style with no text in it", "quality": "hd"}]

Expected Behavior:
Prevent the text relating to tool calls to be displayed after the prompt when waiting for users input.

@sangee2004
Copy link
Contributor Author

Tested with gptscript version v0.0.0-dev-463fe293-dirty

This issue is not seen anymore.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants