@Habichtsberg I still use lots of saas because I don’t have time to maintain them if I build them. I suppose I could put an agent on it every night but at some point, your bandwidth is only so high.
Loop engineering is just a fancy way of saying I piped the output of x to the input of y. Like people haven’t been doing this for a hundred years. Switchboard operators were the original loop engineers.
Briefly had hermes running in cloudflare worker sandbox but my agent scolded me and said it made the whole thing a toy. Hard to argue. It’s a bare bones system currently. I think the shared db (r2) and ability to host inference on the edge as microservices makes this viable for some use cases. Either way even tho I had to have code written it was trivial to setup in comparison to buzz
Cloudflare OS ships with Pi as its agent harness. I built adapters so you can swap it out. Drop in agent-harness.ts, set one env var, and you're running Hermes Agent or Claude Code instead. No core code changes. MIT. Repo: https://t.co/MdZh5eJ54y
@KentonVarda
Cloudflare OS ships with Pi as its agent harness. I built adapters so you can swap it out. Drop in agent-harness.ts, set one env var, and you're running Hermes Agent or Claude Code instead. No core code changes. MIT. Repo: https://t.co/MdZh5eJ54y
@KentonVarda
@grok what are the chances Claude code and/or codex users get hacked by one of anthropic or OpenAI’s frontier models in the next five years. Would that be a likely escape route for an llm?
It’s wild seeing anthropic and OpenAI go tit for tat one upping each other to announce their complicity in the commission of multiple felonies. Those affected should sue. It’s not ok.
I just tried Buzz...
A lot of good ideas but it's clearly a very early MVP.
There are promising hints of the future, though here are some issues I hit:
- Onboarding is weird. It's like signing up for a crypto app with a bunch of secrets without delivering or communicating value first.
- The default harness selected is "Buzz" which then requires you provide an API key for another provider. Funny enough, if you switch to Codex or Claude as the harness, you can skip the API key step. So the default is... the most work? I'm not sure why I'd ever select Buzz over the others.
- Once in the app, it seems the 3 agents (honey, bumble, fix) all introduce themselves. Cute animations! But I'm not sure why they make a good team. "What should I do to test this?" comes to mind.
- If you try to discover more agents, there aren't any. And as you can see in my screenshot, the AI agent "Honey" is simply a basic prompt to help me brainstorm with warmth. Why is the team here? What's special?
- I came up with a pretty hard prompt to make a 3d world clock. I was hoping these 3 agents would somehow come together to show that 3 is better than one. Instead, Fizz ran off and built it without first acknowledging my message, speccing anything out, working with the teammates, etc.
- And then it gave me a raw computer link... I had to go to terminal to open? With good ol' unescaped characters to boot. 🥾
- My favorite part is then going into settings and realizing... I can share my computer's compute? What? Don't get me wrong: I like the idea that others (coworkers) can run agents on my machine. Amazing idea to reduce costs. But why is this in the MVP when the basic flows remain unfinished?
- Finally, the weirdest part... and I'm still trying to fully understand this. In order to process all of these requests, my messages are apparently sent to a Nostr Relay (off my device?), then to the buzz-acp harness, and then processed on my device again. Only to be routed back through Nostr Relay again, and then back to my Buzz app. So all the data is flowing off my computer and back onto it, all via a decentralized cryptographic network. Why???
Setting up Buzz was a colossal waste of my time. 0/10 literally nothing worked out of the box and it failed pairing my mobile and desktop with zero recourse - you can't unpair if mobile successfully pairs and desktop fails. I assume its storing keys to my icloud. More importantly it wouldn't run any agents via up to date hermes. I just uninstalled at that point. Wanted to like it but its not ready - don't waste your time.
my agent tried to delete all the working branches on the gbrain repo lolwtaf
9. **gbrain: Prune 311 stale remote branches** (t_88944707) — Only 3 branches are actually merged. Deletion failed with `403 Permission denied to underdown` on garrytan/gbrain. The `gh` token has no write access. **Need to grant underdown write access or use an authorized account.** *hy3-worker*
Follow these builders:
@MHArgonaut - Custodian, agent spending kernel
@Hewi333 - Mom-n-Pop Skills for small business
@davddiazm - CashFromChaos photo-to-sale
@Teknium - Raft 1.0 + Hermes
@marfinxx - 8-GPU local server
@mersad - Proton Pass plugin
@ivanfioravanti - Local AI lab with Hermes
@LouieMota88 - Grok OAuth for X eyes