-
Notifications
You must be signed in to change notification settings - Fork 5.2k
Closed
Description
Is your feature request related to a problem? Please describe.
First of all many thanks to anybody involved in this project, I really appreciate your time and effort.
Shouldn't a command (such as "git") asking for stdin be aborted by the LLM model in use and propose an alternative?
I am running (not sure if relevant) dockerized version of open interpreter docker-open-interpreter:
docker run \
-it \
--rm \
--name agent \
-e DEEPSEEK_API_KEY=$DEEPSEEK_API_KEY \
-v /var/run/docker.sock:/var/run/docker.sock \
ghcr.io/legido-ai/docker-open-interpreter interpreter \
--model deepseek/deepseek-chat \
-y --safe_mode off
And I asked:
Clone https://github.com/legido-ai/private
I'll clone the repository from GitHub. Let me first check the current directory and then clone the repository.
pwd && ls -la
/workspace
total 8
drwxr-xr-x 2 appuser appuser 4096 Dec 1 17:16 .
drwxr-xr-x 1 root root 4096 Dec 2 16:17 ..
Now I'll clone the repository:
git clone https://github.com/legido-ai/private.git
Cloning into 'private'...
Username for 'https://github.com':
Not sure if it's the model (deepseek/deepseek-chat) or simply I'm not understanding how open-interpreter works, but I expect the model to halt and follow another strategy, like for instance claude code will do.
Thanks for your feedback!
Describe the solution you'd like
Add a flag to consider stdin as an error, so model can replan
Describe alternatives you've considered
No response
Additional context
No response
endolith
Metadata
Metadata
Assignees
Labels
No labels