Composal

Key facts: Composal tests your product the way users do. Give it your URL and connect your repo, specs, PostHog or Sentry, and it sends a group of browser agents through your key flows and edge cases at the same time. When something breaks, you get a video, logs and steps to reproduce it. Pass those to your coding agent through Composal's CLI, and Composal checks the fix. It is part of Y Combinator's Fall 2026 batch.

Plans: Pricing is not listed on the site. You can get started from the site or book a call.

Best for: Teams shipping fast with coding agents who need proof their app still works. Who should skip it: tiny projects with one flow that manual testing already covers.

Pros

• Tests with a swarm of browser agents running real user flows and edge cases
• Every failure comes with video, logs and reproduction steps
• Its CLI hands the evidence to your coding agent and then re-checks the fix
• Connects to repos, specs, PostHog and Sentry for context

Watch out for

• Pricing is not listed, so budgeting needs a call or signup
• A new Y Combinator product, so coverage of unusual flows is still maturing
• Agent testing complements, but does not fully replace, human QA on critical paths

Best for

Teams shipping fast with coding agents who need proof their app still works. Who should skip it: tiny projects where manual testing already covers the one flow that matters.

Pricing

Pricing is not listed on the site. Get started from the site or book a call for details.

FAQs

How is Composal different from a test suite? It sends browser agents through your product like real users, across key flows and edge cases at the same time.

What do I get when something breaks? A video, logs and steps to reproduce it.

Can it check the fix? Yes, pass the evidence to your coding agent through its CLI and Composal verifies the fix.

What can it connect to? Your repo, specs, PostHog and Sentry.