Proof Docs
Open Proof
Autonomous product QA

Introduction

Proof learns the customer outcomes your product must protect, exercises them in a real browser, and turns successful AI exploration into deterministic Playwright coverage.

What Proof does

Proof starts from a business task, not a selector list. It observes the page, chooses browser actions from the current state, verifies the resulting customer outcome independently, and compiles the verified semantic trace into a Playwright test that runs without AI.

When that deterministic test later fails, Proof retries the original business purpose. If the outcome still works, it can prepare a validated test repair. If the outcome no longer works, it reports a regression with the run history and captured evidence.

Ways to use Proof

Slack
Ask Proof to check a customer task from Slack.

Every entry point creates the same durable run and returns the same business outcome.

Read the guide
Command line
Configure and run Proof from a terminal.

Every entry point creates the same durable run and returns the same business outcome.

Read the guide
GitHub pull requests
Check affected user flows automatically.

Every entry point creates the same durable run and returns the same business outcome.

Read the guide
Production monitoring
Run saved tests in production without AI.

Every entry point creates the same durable run and returns the same business outcome.

Read the guide

The mental model

  1. Project context: the approved Markdown description of your product.
  2. User flow: a short description of an important feature or journey.
  3. Test: a generated browser check derived from an approved user flow.
  4. Exploration: an AI agent reasons through the check in a browser.
  5. Semantic trace: the verified, selector-independent record of what worked.
  6. Compiled test: a versioned Playwright test replayed without model calls.
  7. Adjudication: a fresh semantic attempt that decides whether a failed test is stale or the product is broken.