Composal
Composal provides an automated testing platform that uses AI browser agents to pressure-test critical user flows and edge cases before release. The platform maps product features through a Discovery agent, runs parallel testers via Deep Scan, and converts findings into repeatable plain-English scenarios that run on every pull request. Teams can send bugs with video evidence and reproduction steps directly to coding agents like Claude, Codex, or Cursor for rapid fixes.
- Developer Tools
- AI Agents
- Artificial Intelligence
- Software Only
Funding
Raised to date
$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Founder details are not available yet.
Product
Problem
Development teams struggle to verify that critical user flows work correctly before each release, relying on manual testing that is time-consuming and often misses edge-case bugs. This slows down shipping and allows regressions to slip into production.
Solution
Composal provides an automated testing platform that maps a product's features and user flows, then deploys a swarm of AI agents to pressure-test them. The Discovery agent reads product specs to build a persistent knowledge map of how the product should work, which agents use to explore areas and identify bugs. Deep Scan launches parallel testers across user flows, capturing video recordings, technical evidence, and reproducible steps for each issue found. Teams can convert findings into repeatable scenarios written in plain English, run them on every pull request from CI, and send bugs with full evidence directly to coding agents like Claude, Codex, or Cursor for rapid fixes.
Target Audience
Primary users are software development teams and engineering organizations that need to verify critical user flows before release, particularly those using AI coding agents and looking to reduce manual QA effort.
Features
- Discovery agent that reads product specs and maps features, user flows, and areas of the product into a persistent knowledge base
- Deep Scan that deploys a swarm of parallel testers (e.g., 10 browsers, 20 testers) to pressure-test flows and find edge-case bugs
- Automatic generation of reproduction steps, video recordings, and technical evidence for each discovered issue
- Plain-English scenario authoring that turns observed journeys into repeatable regression tests
- CLI tool that lets developers run scenarios, launch scans, and send bugs with evidence directly to coding agents like Claude, Codex, or Cursor
- Scenario relevance scoring that automatically selects and runs the right tests for each pull request
- Knowledge documents that teams can review and edit, with overnight refinement of what the system learned