Abyss проти ScreenCI

Stop trusting your AI agent with your machine. Give it a sandbox it can't escape. [Abyss](https://abyss.scurry.io/) is my take on a local-first agent containerization framework for ACP agents. It aims to neatly solve 2 issues:1. Agents having full access to your desktop, and thus indiscriminate access to your files/secrets/etc.2. Agent/harness features having to be re-implemented for each agent, even if the feature isn't tied to the specific agent in any way.The former is largely self-explanatory, I think. I want to log into my bank, I don't want my agent to even have the ability. There's a fair bit of software in this space a...

Tutorial videos of your app, made by your coding agent. Describe the feature in one sentence and paste one prompt into Claude Code, Codex, or Cursor. It records a narrated walkthrough of your real app on your own machine, and the finished video opens in your browser, ready for your help center or docs, where anyone edits the narration, voices, and 79 languages. Engineers can keep the videos in the repository as Playwright tests that CI re-records on every release. Hi HN, I'm Olli. My friend Patrik and I are bootstrapping ScreenCI: tutorial videos of your web app that you can re-record in your CI on every release. If the video is out of date, the run fails like any E2E test.I've been on teams where the tutorial videos were outdated, or even where we held back UI improvements due to manually maintained videos. That's why we built system where you can describe the videos as Playwright E2E test code. Coding agents are very good with writing tests and you can run these scripts against your app on your machine. ScreenCI converts them into narra...