Best AI tools for Automated Testing & QA
19 AI tools on FyndMyAI™ carry the catalog tag Automated Testing & QA. This page links 12 with the longest catalog descriptions. It is not a scored
19 AI tools on FyndMyAI™ carry the catalog tag Automated Testing & QA. This page links 12 with the longest catalog descriptions. It is not a scored ranking. Including evalship, Regresco, Kerno.
These are not ranked by reviews or price. All Software & Development tools. All best hubs.
Tools
- evalship — A GitHub App that reviews LLM evals and comments on pull requests. It flags evals that pass when they should fail, reports what a PR broke
- Regresco — Saved browser flows for a staging site, run in Chromium through Playwright. You give it a URL, describe a flow in plain text, accept flows
- Kerno — Checks backend changes inside a coding-agent loop, and keeps a code graph called KIT. Community is $0 for 30 test runs a month, one repo
- UI Verify — Screenshots the screens a pull request changes in Playwright, Storybook, or Vitest, and an AI judge separates an intended change from a
- Zof — Sits between an AI change and production. It explains the change, checks behavior, collects evidence, and can block a risky release
- Tiden — Stores what you asked a coding agent to build as versioned specs, then checks each change against those specs before it ships. Starter is
- AegisRunner — AegisRunner takes a website address or an uploaded Android or iPhone app, writes the tests without code, and runs them on real browsers and
- Noveum.ai — Records instrumented agent runs and scores chat, voice, and workflow traces against a library of published checks. Free includes 2,500
- Confident AI — Tests LLM apps with research-backed metrics, traces production traffic, and red-teams them before they ship. The free plan includes 5 test
- PromptJury — PromptJury runs one set of test cases through several prompts and models, then has other models score the answers without seeing who wrote
- LaunchProof — Opens your site or app on iPhone, Android, and major desktop browsers, then returns one report that names each problem and a fix. A smaller
- Screenery — Screenery publishes screenshots a project's own tests already capture, each at a stable URL that follows the default branch, and the only