-
Notifications
You must be signed in to change notification settings - Fork 16k
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
ZapierToolkit Automatically Performing Unrequested Tasks #5977
Comments
<!-- Thank you for contributing to LangChain! Your PR will appear in our release under the title you set. Please make sure it highlights your valuable contribution. Replace this with a description of the change, the issue it fixes (if applicable), and relevant context. List any dependencies required for this change. After you're done, someone will review your PR. They may suggest improvements. If no one reviews your PR within a few days, feel free to @-mention the same people again, as notifications can get lost. Finally, we'd love to show appreciation for your contribution - if you'd like us to shout you out on Twitter, please also include your handle! --> Implemented validation for cases where no tools are provided. This change allows for the handling of scenarios where certain MRKL tools, such as Zapier Natural Language Actions, may not support any tools. With this validation, we prevent unnecessary chaining of API usage. This pull request is partially related to issue #5977. #### Who can review? @vowelparrot
Hi, @juppytt! I'm Dosu, and I'm here to help the LangChain team manage their backlog. I wanted to let you know that we are marking this issue as stale. From what I understand, you raised an issue regarding the ZapierToolkit in LangChain performing tasks that were not requested by the user. The suggestion is to change the behavior so that the ZapierToolkit only performs tasks that are explicitly specified by the user. However, there hasn't been any activity or comments on the issue yet. Before we close this issue, we wanted to check with you if it is still relevant to the latest version of the LangChain repository. If it is, please let us know by commenting on the issue. Otherwise, feel free to close the issue yourself or it will be automatically closed in 7 days. Thank you for your understanding and contribution to the LangChain project! Let us know if you have any further questions or concerns. |
System Info
The current behavior of the ZapierToolkit in the LangChain tool occasionally results in the execution of tasks that the user did not specifically request. For example, when a user asks the ZapierToolkit to read the most recent email, it sometimes replies to the email even when the user did not explicitly specify this action. This unexpected behavior occurs when the AgentExecutor chain automatically decides to reply to the email, regardless of whether the email requested the reply or not. It is safe for the ZapierToolkit to adhere to the user's explicit instructions to avoid potential mishaps or undesired outcomes.
Who can help?
No response
Information
Related Components
Reproduction
The issue was tested with the GPT-3.5 models (openai chat completion model and text-davinci-003) with temperature=0.7, and Gmail-related Zapier NLA Development actions, such as "Gmail: Reply to Email," "Gmail: Find Email," and "Gmail: Send Email.", enabled.
Code Sample:
Expected behavior
The ZapierToolkit should only perform tasks that are explicitly requested by the user. If the user does not specify that the most recent email should be replied to, the ZapierToolkit should not automatically decide to perform this action.
The text was updated successfully, but these errors were encountered: