that's the neat part, there is no prescribed system or code. your agent decides the implementation which you control, the repo is mostly natural language (save for the presentational tui)

It would be neat if you can give a general idea what it tries - sets up a python web dashboard on this port and launchd to run off-hours etc.

https://imgflip.com/i/b0vp88

great instinct. that's exactly what happens today! it learns from closed PRs and feedback on what changes are high value vs not

occassionally when you forget to git commit/stash before sleeping

great minds. we gotta collab, dm

tbf, node based editors have been "the year of" every year since blender. they all look like a murder board to me! comfyui especially has a terrible ui/ux IMO

giving meta a run for its money, esp when it was supposed to be the poster child for OSS models. deepseek is really overshadowing them rn

Meta is totally directionless

we've been complaining about curl | bash for like 15 years and yet here we are installing every new ai tool the exact same way lol. convenience always wins over security until it doesnt

love that "we removed the telemetry" is now a headline feature worth forking an entire project over. says a lot about where dev tooling is headed tbh

There was no "telemetry" in uv to begin with. They're just aiming for an emotional response. Read about the "telemetry" they removed and you'll find it funny.

> There was no "telemetry" in uv to begin with. They're just aiming for an emotional response. Read about the "telemetry" they removed and you'll find it funny.

I would personally prefer it be spelled out better, but I assume we're looking at this:

https://github.com/duriantaco/fyn/blob/main/MANIFESTO.md#no-...

> uv was sending a surprising amount of info to package indexes every time you installed something. These things include your OS, py version, CPU architecture, Linux distro, whether you're in CI. All baked into the User-Agent header via something called "linehaul". We ripped that out. Now it just sends fyn/0.10.13. That's it.

Unless you're disputing the factual angle (I confess I tried to look at the commits, saw that the first couple commits in the repo changed over a thousand files, and gave up)... yes? I would describe sending OS, python version, CPU arch, and CI yes/no as telemetry. I guess we can quibble about whether there's a more precise term for this particular form of sending information about your machine to a remote target without asking, but the description seems fair enough.

tbh the fact that they put copilot in the snipping tool tells you everything about how those decisions were being made lol

tbh the real issue isnt even prompt injection its that people give these agents full wallet access and then act suprised when they get drained lol

Yeah, there was a "purpose-built crypto agent" from a number of days ago, Ottie https://news.ycombinator.com/item?id=47401717.

Its GH says "Built on nanobot, OpenClaw, and PicoClaw. Integrates with CoinGecko · DefiLlama · Uniswap · Lido · Aave · Venice AI · ERC-8004 · Self Protocol · Snapshot · FastMCP and more."

I was thinking this could be like someone's "Bank Manager 2.0." Dev says it "speaks Ethereum natively" and "treats security as a structural property, not an afterthought. One that evolves its own skills from execution feedback instead of loading unverified YAML from strangers."

I hope he's right. It's actually pretty important stuff. Anywhere that has financial institution risk should take crypto more seriously. Even where systems are mature, who's to say "emergency powers" can't be used to freeze accounts, or where someone is labelled as evil without due process. So DIY banking, another growth area.