Get started

Testsy docs

Describe what a tester should do and check, in plain English. Testsy runs it in a real Chrome, or on iPhone and Android apps, and comes back with the verdict, the value behind every check, a replay and a fix prompt.

Start here#

Quickstart: your first test and its result in about five minutes.
How a run works: what happens between "Run" and the verdict.
Write a test with the assistant: describe it, review the plan, save.

Home: what needs attention, what is running now, and your projects.

Read results#

Verdicts and what counts: passed, failed, blocked, inconclusive, did not finish.
Proof, replay and video: the value behind every check, where it was read, and the recording.
Fix prompt and share links: hand a failure to a coding agent or a client.

Automate#

Run tests in CI: GitHub Actions, GitLab CI or any CI, failing the build on a failed check.
AI agents (MCP): let Claude Code, Cursor or Codex write tests, run them and read the results.
API reference: every action the app has with tests and runs, over HTTPS.

From the command line#

Terminal

# 1. Create a key in Settings → API keys, then:
export QA_AI_API_KEY=qa_…
curl -fsSL https://app.testsy.dev/cli/qa-ai.mjs -o qa-ai.mjs

# 2. Run every test in a project and wait for the verdicts
node qa-ai.mjs run --project p_your_project