← All IntelClip / AI ToolsSteering the agent: one bash script instead of spamming GitHub clones
From Tencent HY3 · ≈15:22
Demonstrates a reusable human-in-the-loop technique: when an agent flails cloning random repos, redirect it to gather system facts into a reference file it can reason over.
What’s in it
- Demonstrates a reusable human-in-the-loop technique: when an agent flails cloning random repos, redirect it to gather system facts into a reference file it can reason over.
Clip transcript
do this, so we'll see however it goes about doing that. All righty, it keeps just trying to clone repos that it's not fine. I'm going to stop at this for a second because I don't like what it's doing. Why don't you All right, so I'm telling it to write a simple bash script that I will manually run and type the root password so I can get the pertinent information from the system it needs and then save that info to a text file so it can reference that and understand what the specific device has so it can try to reverse engineer whatever protocol is being used for RGB as opposed to this is just basically like trying to spam download a bunch of random GitHub stuff and most of it's not working. So I'm kind of giving it a little bit of a a just like a you know, a tip. All
Comments
Sign in to comment.
Loading comments…