←back to thread

The unreasonable effectiveness of an LLM agent loop with tool use

(sketch.dev)
435 points crawshaw | 2 comments | 15 May 25 19:33 UTC | HN request time: 0.398s | source
Show context
mtaras ◴[16 May 25 07:33 UTC] No.44002704[source]▶
Just a couple of days ago I discovered this truth myself while building a proactive personal assistant. It boiled down to just giving it access to managing notes and messaging me, and calling it periodically with chat history and it's notes provided. It's surprisingly intelligent and helpful, even though I'm using model that's far from being SOTA (Gemini Flash 2.5)
replies(1): >>44003114 #
dandaka ◴[16 May 25 08:44 UTC] No.44003114[source]▶
Would love to learn more! Managing notes — is it your already existing docs? I have been thinking about proactive assistants, but don't really know where to start. I have a few product ideas around that, where this proactivity can deliver a lot of value.

My personal experience with building such agents is kinda frustrating so far. But I was only vibe coding for a small amount of time, maybe I need to invest more.

replies(1): >>44004504 #
mtaras ◴[16 May 25 12:16 UTC] No.44004504[source]▶
No, I personally don't have notes that would be valuable enough for the system I was looking for, so notes is just how I'm calling my agent's long-term memory. All I do is provide it with tools to message me and manage it's own notes, useful context (recent chat history, its saved notes, calendar events, etc), and it can act upon the info.

The elegance of the system unfolded when I realized that I can not specify any interaction rules beforehand — I just talk to the system, it saves notes for itself, and later acts upon them. I've only started testing it, but so far it's been working as intended.

replies(1): >>44005858 #
1. triangleman ◴[16 May 25 14:20 UTC] No.44005858[source]▶
What do you mean by "message me" and "later" in this context?
replies(1): >>44005930 #
2. mtaras ◴[16 May 25 14:27 UTC] No.44005930[source]▶
"Message me" is through a bot in my main messenger app, so the agent is just a chat window for me. "Later" happens per my requests, and agent has an ability to do something later because I just automatically invoke it on a schedule periodically, and ask it "check notes and chat history — do you need to do something, like message the user to remind them of something", and agent has the message_user tool to do that (that goes to the messenger bot API). Or it can decide to do nothing.