What it does Ui.Vision is a browser extension for Chrome, Edge, and Firefox that records and replays web automation macros. It imports and exports Selenium IDE scripts, and the project description claims computer vision, OCR, and Anthropic Claude integration — though the README itself never explains how any of that actually works.

The interesting bit The Selenium IDE compatibility is the concrete hook. Most RPA tools abandoned that format years ago; Ui.Vision still treats it as a first-class citizen, which matters if you’re sitting on old test suites or iMacros scripts you never migrated.

Key highlights

• Free for private and commercial use (no tier mentioned, no catch stated)
• Builds for three browser engines from one Node codebase (dist for Chromium, dist_ff for Firefox)
• Node 20.11.1 / NPM 10.2.4 toolchain — refreshingly specific
• Distribution through official web stores, not just sideloading
• Active forum support rather than GitHub issues as primary channel

Caveats

• The README is essentially an install guide; architecture, security model, and the touted AI features are unexplained
• No screenshots, no demo GIFs, no command reference in-repo — everything points outward to ui.vision
• Build instructions assume you already know why you’d want to compile a browser extension from source

Verdict Worth a look if you need to resurrect or extend Selenium IDE workflows without paying for cloud RPA. Skip it if you want transparency about how the “AI” and “computer vision” pieces actually function — that story lives elsewhere, if anywhere.