opslane

opslane / verify

Public

Verification Layer for Claude Code

claude claude-code claude-code-plugins claude-code-skill
80
3
100% credibility
Found Mar 11, 2026 at 65 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

This tool automates checking web applications against a specification document by running AI-driven browser tests on a local development server and generating reports with screenshots and video recordings.

How It Works

1
🔍 Discover the checker

While building your web app, you hear about a smart tool that tests if it matches your plans before sharing.

2
📥 Add it to your toolkit

You easily add this verification helper to your AI coding assistant with a simple command.

3
🔑 Handle app login

You set up login details for your app once, so tests can act like real users.

4
📄 Share your plans

You provide the document listing what your app should do, and it asks any quick questions to get started.

5
🤖 Watch it test

It automatically opens browsers to check each requirement on your running app, taking screenshots and recordings.

6
📊 Get the report

You receive a clear summary with pass or fail for every check, plus pictures and videos to see exactly what happened.

✅ Push with confidence

Now you know your app works as planned and can safely share your changes.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 65 to 80 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is verify?

Verify is a shell-based plugin for Claude Code that automates acceptance testing on your local dev server. Feed it a spec doc outlining acceptance criteria, and it spins up one AI agent per criterion using Claude and Playwright to interact with your app, capture screenshots, and record sessions. You get pass/fail results with evidence before pushing code—no CI pipelines or remote infra required.

Why is it gaining traction?

It skips the hassle of manual e2e tests or setting up CI by running everything locally via simple commands like `/verify-setup` and `/verify`. Developers dig the instant feedback with clickable HTML reports, trace viewers, and session videos right in `.verify/evidence`, making debugging "filesystem layer verification failed" errors or UI regressions dead simple. The Claude integration feels native if you're already using it for code gen.

Who should use this?

Frontend devs prototyping web apps with Claude Code who hate writing Playwright tests by hand. Solo hackers or small teams verifying features against spec docs locally, especially before PRs to dodge "github verification code not working" surprises. Anyone tired of "docker pull" hangs from untested layers on CI.

Verdict

Worth a spin for Claude Code users—42 stars and 1.0% credibility score signal early days with thin docs, but solid README and zero-setup install make it low-risk to try. Pair it with your dev workflow for quick wins until it matures.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.