RPA
Ui.Vision is a browser extension for Chrome, Edge and Firefox that automates websites and, with extra components, desktop applications. It combines a macro recorder and command table, a JavaScript API (`uiv.*`), OCR and image recognition. The README lists browser tasks (form filling, data extraction, downloads), website testing with Selenium IDE commands, visual tasks that find on-screen text or locate controls by image, and desktop workflows using visual recognition and mouse and keyboard input, which need the additional XModules components.
For AI use it provides a Model Context Protocol server through the `uivision-mcp-bridge` package, connecting clients such as Claude Code, Claude Desktop and Cursor to the extension. An agent can inspect a page, create or edit a macro, run it and check the result from logs and screenshots, and the resulting macro is saved to run again. Setup is `npx uivision-mcp-bridge --setup`, enabling the bridge in Ui.Vision's AI settings and restarting the client; there is also a built-in AI assistant, and the README links an AI-oriented API reference and a plain-text `llms-full.txt`.
You install it from the Chrome Web Store, Edge Add-ons or Firefox Add-ons; building from source (Node 20.11.1) is only for developers. The README says the browser extension is open source and free for personal and commercial use, with desktop components and editions described on the vendor's site. GitHub did not detect an SPDX license for the repository. It was last pushed on 2026-10-05. This review is based on the README and repository data; we have not run it ourselves.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-09-21
- Commits, last 90 days
- 2
- Releases, last 90 days
- 1
- Latest release
- V10.0.276 (2026-09-21)
- Open issues
- 15
- Recent issues with at least one comment
- 10/15
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #162 Feature request: Support custom URL for anthropic upstreams.0 comments · 2026-11-06
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Who it fits
- You want macros that people can record, edit and rerun in Chrome, Edge or Firefox, with an AI assistant writing and running them through MCP.
- You need desktop automation or image and OCR matching as well as browser steps; the README says desktop features need the additional XModules components.
- You prefer reusable saved macros to one-off agent actions.
- You want a plain MCP server that starts with one npx line: setup is an installer plus a browser extension and enabling the bridge in Settings.
- You need a standard OSI licence: the repository's licence is detected as NOASSERTION, and the README describes desktop components and editions separately.
- You only need a lightweight browser driver for an agent; Playwright MCP is simpler.
How it differs
Playwright MCP lets an agent act on a page directly. Ui.Vision lets an agent create and edit macros that you then run and inspect in the extension, which suits repeatable RPA workflows rather than ad-hoc browsing.
Compatibility
Quick start
Use cases
Alternatives
Featured in these lists
Comments
No comments yet. Be the first!