Releases·Tool Release
pixelpact: GitHub tool extracts UI contracts from reference pages and gives coding agents numeric drift feedback
pixelpact, released on GitHub, provides measurement for coding agents building UIs. It extracts a contract from a reference page, checks the implementation against it, and returns numeric scores. This addresses the problem that coding agents cannot see the page they build.
24 engagement1 sourceSat, Sep 5, 2026, 09:09 PM
AI Overview
- pixelpact (GitHub, 2026-09-05) extracts a contract from a reference page — sizes, colors, spacing, typography, hover/focus states, animation keyframes, design tokens
- `pixelpact check` reports one line per drifted property, closing the loop for agents like Claude Code and Cursor via stdio MCP, Playwright, and GitHub Actions
Entities
Related
Products·launched
Pinpoint: Visual feedback tool for AI coding agents
Pinpoint is a tool that lets humans give visual feedback to AI coding agents by dropping contextual comments on screenshots. It solves the problem of typing out feedback for autonomous agents, making the human-in-the-loop process more natural and efficient. It also supports side-by-side before/after comparisons.
Pinpoint is a tool that lets humans give visual feedback to AI coding agents by dropping contextual comments on screenshots. It solves the problem of typing out feedback for autonomous agents, making the human-in-the-loop process more natural and efficient. It also supports side-by-side before/after comparisons.
3 engagement1 sourcehackernewsTue, Jul 21, 2026, 01:12 PM
Products·launched
GridProof: MCP server that audits and fixes spacing/grid drift in AI-generated UIs
GridProof is an MCP server that integrates into the AI coding loop to catch and fix spacing and grid inconsistencies in generated frontends. It uses Playwright to render the page, measures computed geometry, and returns structured fix hints for the agent to apply, then re-audits. It solves the problem of subtle UI drift that ships unnoticed, for developers using AI agents like Claude.
GridProof is an MCP server that integrates into the AI coding loop to catch and fix spacing and grid inconsistencies in generated frontends. It uses Playwright to render the page, measures computed geometry, and returns structured fix hints for the agent to apply, then re-audits. It solves the problem of subtle UI drift that ships unnoticed, for developers using AI agents like Claude.
3 engagement1 sourceredditFri, Aug 14, 2026, 12:52 PM
Social·Community
Users share frontend verification workflows between Claude Code 'done' and git commit
A developer asks how others verify frontend changes made by Claude Code before committing, noting they no longer trust the agent's self-verification. They built a local tool that snapshots pages before and after agent work to produce a proof sheet of visual changes.
A developer asks how others verify frontend changes made by Claude Code before committing, noting they no longer trust the agent's self-verification. They built a local tool that snapshots pages before and after agent work to produce a proof sheet of visual changes.
12 engagement1 sourceredditFri, Aug 14, 2026, 03:23 PM
Products·launched
Coding agent that sees browser to iterate on UI changes
Peek CLI is a tool that lets coding agents observe the browser in real time, enabling them to iterate on UI changes by seeing the visual output of their code. It solves the problem of agents working blind on frontend tasks, making them more effective for developers who want automated UI development.
Peek CLI is a tool that lets coding agents observe the browser in real time, enabling them to iterate on UI changes by seeing the visual output of their code. It solves the problem of agents working blind on frontend tasks, making them more effective for developers who want automated UI development.
5 engagement1 sourcehackernewsMon, Jul 13, 2026, 12:07 AM
Products·launched
CodeVetter: Local verification for coding agent output
CodeVetter is a desktop app that verifies code changes made by coding agents locally. It collects the task, agent changes, executable checks, and evidence into a review bundle, helping developers confirm that requested behavior actually works without relying on a second model. It targets TypeScript and Node apps with browser or API behavior.
CodeVetter is a desktop app that verifies code changes made by coding agents locally. It collects the task, agent changes, executable checks, and evidence into a review bundle, helping developers confirm that requested behavior actually works without relying on a second model. It targets TypeScript and Node apps with browser or API behavior.
4 engagement1 sourceredditSun, Aug 16, 2026, 10:03 AM