Viewing profile — tonyww

Browser use is a token hog

[dead]

If you are building AI agents with frameworks like browser-use, LangChain, or OpenClaw, you've likely hit the "blast radius" problem. A misconfigured prompt or hallucination can ca…

AI agents currently operate on a flawed security model: they inherit the ambient permissions of the terminal they are spawned in. If an agent gets prompt-injected or hallucinates, …

Yeah, that’s a pretty good analogy. The main difference is that the “tests” are predicates over live browser state and are often proposed alongside the plan on the fly, not written…

Absolutely agree on the compounding error point - that’s exactly what pushed us toward verification. On “verification wrong”: we try hard to keep predicates grounded and re-evaluat…

It’s mostly the former: there’s a small set of generic checks/primitives, and we choose which ones to apply per step. The binding between “task/step” and “what to verify” can come …

I’m absolutely not AI, I dedicate this morning to technical discussion with HN community on my post, which I’ve spent weeks building the technology behind it

Totally agree - hybrid approaches can work well, especially on messy pages. We’ve seen the same tradeoff. On the verification side though, dynamic pages are exactly the reason why …

Importance ranking is just a heuristic pass that scores/prioritizes elements (size, visibility, role, state) so the snapshot stays small and focused. It’s deterministic, not ML. Th…

The WASM pass is fully deterministic: it’s just code running in the page to extract and prune post-rendered elements (roles, geometry, visibility, layout, etc), no agent involved i…

Thanks — that’s exactly our motivation. The key shift for us was moving from “did the agent probably do the right thing?” to “can we prove the state we expected actually holds.” Th…

The accessibility tree is definitely useful, and we do look at it. The issue we ran into is that it’s optimized for assistive consumption, not for action verification or layout rea…

A quick clarification on intent, since “browser automation” means different things to different people: This isn’t about making scripts smarter or replacing Playwright/Selenium. Th…

A common approach to automating Amazon shopping or similar complex websites is to reach for large cloud models (often vision-capable). I wanted to test a contradiction: can a ~3B p…

Sorry for the misunderstanding, I intended to post it as news or engineering article, which is why I didn't include *Show HN* in the title

Good question. On the surface, it does look very similar to the traditional scraper/script, but there's a subtle difference in where the logic lives and how failures are handled. A…