chw9e
Re:Your sign-up form is a weapon
on Thursday April 02, 2026 @02:50AM (#47610829)
This happened to me several years ago. I got signed up to probably 700 newsletters overnight. In the middle of all of the sign ups there was activity on my airbnb account where my notification settings were changed. when i checked my airbnb i noticed that someone had created a fake listing under my account and disabled booking notifications for it. a real multi-layer scam where the hacker would be making money off a fake listing on someone else's account who would probably never even realize it.
Agent-written tests can't verify agent-written code
Diffusion Models for ARC-AGI: A Retrospective
AI Agents are shipping code faster than we can test
Building iOS UI with Coding Agents Is Slow
Show HN: qckfx – Stop manually checking if your AI agent broke your iOS app
AI coding agents can build iOS UI, but they can't verify it. They can update a screen, but the UI might drift or break, and nobody catches it until a human checks.qckfx gives your agent a baseline. Record a simulator session once. Every tap, scroll, and network response gets captured. On replay, each screen is compared against the original.With our MCP, your agent triggers the tests and gets...
A Guide to Hiring the Illegible
Giving Your Agent Eyes Is Not Enough
Show HN: qckfx – Record your iOS simulator sessions, replay them as tests
I built this because I hate writing tests but I also hate shipping bugs.qckfx records what you're already doing. You tap through the simulator to check your work, and it replays those sessions to catch visual regressions and crashes. No SDK, no code changes, no test files.How it works: finish a feature, click around in the simulator, hit Cmd+Shift+S. That's your test. Cmd+Shift+T to run it...
We Use API Agents to Build Integrations Fast
What I Wish I Could Tell Every Engineer Joining a New Team
Dropbox: Knowledge Graphs, Prompt Optimizers, and MCPs
Re:Launch HN: Phind 3 (YC S22) – Every answer is a mini-app
on Wednesday December 03, 2025 @08:40PM (#46142761)
As someone who never used Phind, how was it better than copilot? Did it integrate data from broader sources or just do a better job of presenting it? Was it faster?
Bug Fixing Is an ETL Problem
Ask HN: How do you handle oncall at 15-30 engineers?
I'm doing research on how oncall breaks down as engineering teams scale, specifically at Series A/B stage companies (roughly 15-30 engineers).The hypothesis I'm exploring: oncall workflows were designed for small teams where everyone knows the codebase. They break at scale because of specialization and context loss before volume becomes a problem.Some patterns I've heard so far:- Oncall...
Finding and fixing software bugs automatically with SapFix and Sapienz
AI-Powered Stagehand + bisect:Finding and Fixing the Commit That Broke Your Code
Agentic Bug Reproduction for Effective Automated Program Repair at Google
A Guide to Clustering LLM Chat Transcripts with Hdbscan
Thoughts on Evals
What Real LLM Monitoring Looks Like (and Why It's Hard)
We're debugging LLMs in production by reading chat logs
Ask HN: After 5 years, why is nobody using VSCode LiveShare?
VSCode has LiveShare, Zed's whole thing is supposedly collaboration, there's a bunch of startups trying to crack this. And yet here we all are, tabbing between our editor and Slack every 5 minutes. Not to mention the constant stream of notifications from Linear, GitHub comments, whatever else.This seems like such an obvious problem to solve. We're already collaborating all day but in really...
Re:Show HN: Browser AI – Use Playwright Without Blowing Up Your Contex...
on Tuesday July 22, 2025 @11:03AM (#44647869)
Oh, I just picked Anthropic’s Sonnet4 to start because it’s really good at agentic stuff and because of the subscription benefits.
I will add more model support soon - any models you particularly want to see?
I’m also hoping to test out RL on the tools to get a fine-tuned model specifically for browser automation eventually.
Re:Show HN: Browser AI – Use Playwright Without Blowing Up Your Contex...
on Tuesday July 22, 2025 @10:51AM (#44647660)
It also has its own cli if you just want to run it alone too
Re:Show HN: Browser AI – Use Playwright Without Blowing Up Your Contex...
on Tuesday July 22, 2025 @10:48AM (#44647620)
Yea it can be, but you will pay Anthropic API price. If you already use Claude code on a subscription then you can reuse your subscription for this MCP, making it effectively free
Show HN: Browser AI – Use Playwright Without Blowing Up Your Context Window
Hey HN! I built Browser AI after getting frustrated with the official Playwright MCP server in Claude Code. If you've used it, you know the pain: - 20+ separate tools that Claude has to juggle (click, fill, navigate, wait_for_selector, etc.) - Context window explosions from verbose screenshot data and DOM snapshots - Constant tool selection errors where Claude picks the wrong tool or...