01
After every feature
Each new story brings acceptance criteria that someone checks step by step, by eye.
PROOFTRACE
An AI tester that proves every promise a feature makes.
Give it a user story. It opens your app in a real browser, does what the story describes, and proves each acceptance criterion with a screenshot of every step. Then it hands back a Gherkin feature and a Playwright test your team can rerun on any environment.
$ prooftrace run --story stories/create-project.md
passed: 8 of 8 criteria · 29.7 s · $0.028
THE GAP
Acceptance criteria are still checked by hand, release after release.
Every feature ships with acceptance criteria, and someone has to prove each one works. That means a test engineer clicking through the app after every deployment, then doing it all again in dev, in UAT and in production. It is slow, it never stops, and it is where most development teams fall behind.
02
On every environment
The same checks run again in dev, in UAT and in production, after every deployment.
03
Scripted by hand
Turning them into automated tests takes days, and the tests drift as the app changes.
Prooftrace automates that work, for a few cents a run.
It reads the acceptance criteria straight from the user story and proves each one in a real browser. Then it hands back two files: a Gherkin feature anyone can read, and a Playwright test anyone can rerun on dev, UAT or production with no AI cost. Stories come from Markdown files and GitHub issues today, with Jira and Azure DevOps next.
HOW IT WORKS
Seven stages, one run.
- 1
Read
loads the story from a Markdown file or a GitHub issue.
- 2
Extract
pulls out each acceptance criterion.
- 3
Plan
matches a saved recipe, or decides to explore.
- 4
Sign in
logs in; the AI never sees the password.
- 5
Execute
drives a real browser, a screenshot per step.
- 6
Emit
writes a Gherkin feature and a Playwright test.
- 7
Verify
runs the tests and marks each criterion proven or not.
WHO ACTS ON A STEP
- Seeda saved recipe step, no AI
- Fallbackthe AI finds an element that moved
- Validatethe AI checks each section
- Exploreno recipe, the AI drives
A REAL RUN
Watch a real run. Click any step.
The interactive replay loads here
Recorded with OpenAI's gpt-4.1 against a sample app. Every screenshot, click and check is real. Press Replay in the console to watch the run again.
TRUST BY DESIGN
No proof, no pass.
A criterion passes only when an assertion linked to it passes in the generated test. Actions alone prove nothing, and a run with an unproven criterion is a failed run.
Attacked in review: nine issues, nine guards
01
A page asks the AI to type the password somewhere else
✓ The password fills only password fields, on the app's own site.
02
The AI is steered to another website
✓ Navigation stays on the app's own site.
03
Page text poses as instructions
✓ Everything on the page is treated as data.
04
A story title smuggles code into generated tests
✓ Every text is encoded; locators are parsed, never pasted.
05
Generated tests try to read your API keys
✓ The test process gets an allowlisted environment.
06
A website rebinds its domain to your local server
✓ The server answers only to its own addresses.
07
A website opens a WebSocket to your local server
✓ Connections from any other site are refused.
08
A project's settings file swaps the browser program
✓ It may only supply the settings the config names.
09
A GitHub reference tries to climb out of its path
✓ It is rejected before any request is made.
NUMBERS
What a run costs.
29.7 s
Create a project: 43 steps, 8 of 8 proven, $0.028
10.7 s
Sign in with no recipe: the AI explored, 3 of 3 proven, $0.017
$0
The same recipe with the AI switched off: 16.3 s, 8 of 8 tests pass, no evidence review
OpenAI gpt-4.1 against the sample app, September 2026.
ARCHITECTURE
One engine, one protocol, adapters for the rest.
CLI
Local server
Console
Engine
ModelsOpenAI, Claude, Ollama
StoriesMarkdown, GitHub issues
BrowserPlaywright
TestsGherkin and Playwright
VerifierPlaywright Test
StoreSQLite
Protocol: typed events and schemas shared by every package
HOW IT WAS BUILT
Designed and built by Abhijeet Solanki.
01
The idea
Acceptance criteria should prove themselves, not wait for someone to click through them.
02
The engine
Seven stages, a typed event stream, and adapters for everything outside it.
03
Real models, real lessons
Live runs changed the design: the AI learned to read page text, to assert evidence, and to fail honestly.
04
Attacked in review
Nine security issues found by probing the code, each fixed and probed again.
05
Launch
Recorded runs you can click through, and code access on request.
18
tasks shipped
320+
automated tests
9
security issues fixed
Like what you see?
CODE ACCESS
Want to read the code?
The source is private. Tell me who you are and I'll send you read-only access on GitHub for 7 days.
Used only to grant access, and deleted after 30 days.
Every criterion proven. Every step on record.