Is there a way to prompt the user with the "human" tool? #9

New Issue

2026-02-15T19:15:16-05:00

yindo commented

2026-02-15 19:15:16 -05:00

Originally created by @Jgmedina95 on GitHub (Aug 15, 2023).

I love this repo, and is very useful. Although im wondering if there's an implementation to prompt the user for the "human" tool. Cause as of now, everything happens in the background, and when the human is "prompted" nothing happens.

Originally created by @Jgmedina95 on GitHub (Aug 15, 2023). I love this repo, and is very useful. Although im wondering if there's an implementation to prompt the user for the "human" tool. Cause as of now, everything happens in the background, and when the human is "prompted" nothing happens.

yindo closed this issue

2026-02-15 19:15:16 -05:00

yindo commented

2026-02-15 19:15:16 -05:00

@sfc-gh-jcarroll commented on GitHub (Sep 13, 2023):

Hey, not sure I understand the question. Want to share a little more details - which app are you running? What's the repro steps and what specific behavior did you see? It doesn't sound familiar from my experience playing with all the apps. Thanks!

@sfc-gh-jcarroll commented on GitHub (Sep 13, 2023): Hey, not sure I understand the question. Want to share a little more details - which app are you running? What's the repro steps and what specific behavior did you see? It doesn't sound familiar from my experience playing with all the apps. Thanks!

yindo commented

2026-02-15 19:15:16 -05:00

@Jgmedina95 commented on GitHub (Sep 13, 2023):

hi! In this particular case, i had an agent that used multiple tools. When trying to use it with streamlit for a nice UI, I was running into the bug that whenever the agent wanted to use the "human" tool from langchain. There was no input pop up for the user, so it seemed that the agent was just thinking continuously. Maybe that was a better description?
sadly I cant share the code as it was part of an internship I was doing.

@Jgmedina95 commented on GitHub (Sep 13, 2023): hi! In this particular case, i had an agent that used multiple tools. When trying to use it with streamlit for a nice UI, I was running into the bug that whenever the agent wanted to use the "human" tool from langchain. There was no input pop up for the user, so it seemed that the agent was just thinking continuously. Maybe that was a better description? sadly I cant share the code as it was part of an internship I was doing.

yindo commented

2026-02-15 19:15:17 -05:00

@sfc-gh-jcarroll commented on GitHub (Sep 13, 2023):

Ah I see - might be a good question for the langchain discord if it's about the agent behavior. Going to close this since it seems like more of a general request vs a particular issue with this repo. Thank you!

@sfc-gh-jcarroll commented on GitHub (Sep 13, 2023): Ah I see - might be a good question for the langchain discord if it's about the agent behavior. Going to close this since it seems like more of a general request vs a particular issue with this repo. Thank you!

yindo commented

2026-02-15 19:15:17 -05:00

@amjadraza commented on GitHub (Apr 30, 2024):

@sfc-gh-jcarroll I am having this issue as well. If you find anything, let me know.

For user input when running agent, I need this tool to have interaction with agent while running the job

@amjadraza commented on GitHub (Apr 30, 2024): @sfc-gh-jcarroll I am having this issue as well. If you find anything, let me know. For user input when running agent, I need this tool to have interaction with agent while running the job

yindo commented

2026-02-15 19:15:17 -05:00