Stop letting your coding agent claim "pixel-perfect". Make it prove 97.49%.

A developer built hwatu, a browser designed for coding agents to verify UI changes via pixel diffing, achieving a 97.49% match in about 87 ms per verification loop. The tool runs headlessly on Linux using WebKitGTK, supports CLI and MCP integration, and aims to replace blind 'pixel-perfect' claims with measurable proof.

Your coding agent edits a component, runs the build, and reports the UI is now "pixel-perfect." It never opened a browser. It's guessing. I got tired of that, so I built hwatu https://github.com/hongnoul/hwatu , a browser designed specifically for the agent verification loop. hwatu --headless http://localhost:3000 open, no focus steal hwatu wait-load block until loaded hwatu eval 'document.title' DOM-level checks hwatu shot /tmp/check.png screenshot hwatu diff --id 1 --baseline before.png pixel diff: 97.49% match + heatmap hwatu close 1 The whole loop, screenshot included, runs in about 87 ms on my laptop. Cheap enough that the agent can verify after every edit instead of claiming success blind. webkitgtk-6.0 . No 170 MB download per project. hwatu mcp , plain CLI, and a 1-line JSON socket. Works with Claude Code, Cursor, Jcode, or any harness.Every benchmark in the repo was measured on real runs with scripts you can rerun: window spawn medians of 13-16 ms, ~87 ms full verification loop, memory measured via PSS across all WebKit child processes. Methodology in docs/benchmarks.md https://github.com/hongnoul/hwatu/blob/main/docs/benchmarks.md . curl -fsSL https://raw.githubusercontent.com/hongnoul/hwatu/main/scripts/install.sh | bash hwatu setup detects Claude Code, Cursor, Jcode, or generic MCP MIT licensed. Linux WebKitGTK . Feedback and issues welcome: https://github.com/hongnoul/hwatu https://github.com/hongnoul/hwatu