AWT (AI Watch Tester)

What is AWT?

AWT is a browser testing tool that writes and fixes its own tests.

You give it your web app's URL. AWT opens a real browser, figures out what's on the page (buttons, forms, links), writes test steps, runs them, and tells you what passed and what failed. If something breaks, the DevQA Loop kicks in — AI reads the error, updates the test or your code, and tries again.

No test code to write. No recording sessions. No manual updates when the UI changes.

Start in 5 Minutes

Option 1 — Cloud (no install, free)

1. Visit https://ai-watch-tester.vercel.app
2. Sign up (email or GitHub — takes 30 seconds)
3. Paste your app URL
4. Watch AWT test your site live

Option 2 — Local CLI (runs on your machine)

# Install (requires Python 3.11+)
pip install aat-devqa
playwright install chromium

# Run the visual dashboard
aat dashboard
# → Opens at http://localhost:9500

# Or test directly from the command line
aat devqa "test the login flow" --url https://your-app.com

That's it. AWT scans your page, writes a test plan, shows it to you for approval, then runs it in a real Chrome window.

How It Works

You give AWT a URL
        │
        ▼
  🔍 SCAN — AWT opens Chrome and reads every button, input, and link
        │
        ▼
  📝 GENERATE — AI writes a step-by-step test plan (you review & approve)
        │
        ▼
  ▶️  RUN — AWT clicks, types, and navigates like a real user
        │
        ├── ✅ All passed → screenshot report saved
        │
        └── ❌ Something failed
                    │
                    ▼
            🔄 DEVQA LOOP — AI reads the failure,
               fixes the test (or your code),
               and tries again (up to 5 times)

The DevQA Loop — AWT's Core Feature

Most testing tools stop when a test fails and wait for a human. AWT keeps going.

When a step fails, AWT:

1. Takes a screenshot of exactly what the browser shows

2. Reads the error message and the visible page content

3. Re-scans the page to check if anything moved or changed

4. Patches the specific failing step and retries

If the failure is a bug in your source code (not just a wrong selector), AWT can trace it — finding the route handler, component, or API endpoint that's misbehaving — and suggest or apply a fix.

# Watch the loop run live
aat devqa "checkout flow test" --url http://localhost:3000

# Or use it with your AI coding tool (Claude Code, Cursor, Copilot...)
# "Test the registration page" → AWT scans, generates, runs, fixes

Four Ways to Use AWT

Agent Skill — Let your AI coding tool drive AWT

# One-line install
npx skills add ksgisang/awt-skill --skill awt -g

# Then ask your AI tool:
"Test the login flow on http://localhost:3000"
"Check if the signup form works"
"Run regression tests after my last commit"
# → AWT scans, generates test steps, runs them, and reports back

MCP Server — Protocol-native

# Add to Claude Code
claude mcp add awt -- python mcp/server.py

# Tools available: aat_run, aat_doctor, aat_list_scenarios, aat_validate, aat_cost

What AWT Is Great At

AWT vs Other Tools

vs Playwright / Cypress

Playwright and Cypress are excellent — and AWT is built on top of Playwright. The difference is who writes the tests:

Use Playwright/Cypress when you want full programmatic control. Use AWT when you want tests without writing them.

vs testRigor

vs Applitools

Applitools specializes in visual regression (pixel-by-pixel screenshot comparison). AWT specializes in functional testing (does the login actually work?). They complement each other — run AWT for functional tests, add Applitools for pixel-perfect visual checks.

Speed & Screenshot Modes

Control the trade-off between thoroughness and speed:

# CI/CD — fastest, minimal storage
aat run --verbosity=concise --screenshots=on-failure scenarios/

# Standard QA — balanced (recommended)
aat run --verbosity=concise --screenshots=before-after scenarios/

# Full audit — every step recorded
aat run --verbosity=detailed --screenshots=all scenarios/

Supported AI Providers

# aat.yaml
ai:
  provider: openai        # openai | anthropic | ollama
  model: gpt-4o
  api_key: ${OPENAI_API_KEY}

Architecture

aat devqa / aat run / aat dashboard
              │
              ▼
    ┌─────────────────────────────────────┐
    │           CLI (Typer)               │
    ├─────────────────────────────────────┤
    │         Core Orchestrator           │
    │  Executor · Comparator · DevQALoop  │
    ├────────────┬──────────┬─────────────┤
    │   Engine   │ Matcher  │  AI Adapter │
    │ web/desktop│ocr/cv/ai │ openai/etc. │
    ├────────────┴──────────┴─────────────┤
    │  Pydantic v2 Models · SQLite Learn  │
    └─────────────────────────────────────┘

All modules follow a plugin registry pattern — add a new engine, matcher, or AI provider by implementing one base class and registering it in __init__.py.

Development

Prerequisites

• Python 3.11+

• Tesseract OCR: brew install tesseract / apt install tesseract-ocr

Commands

git clone https://github.com/ksgisang/AI-Watch-Tester.git
cd AI-Watch-Tester
python -m venv .venv && source .venv/bin/activate
make dev
make test        # verify everything works
aat dashboard    # launch at http://localhost:9500

Contributing

See CONTRIBUTING.md — contributions, bug reports, and new plugins are welcome.

git checkout -b feat/my-feature
make format && make lint && make typecheck && make test
git commit -m "feat(scope): description"

FAQ

No. The Cloud version at ai-watch-tester.vercel.app needs nothing — just a browser. The local CLI needs one terminal command to install.

The only thing AWT needs from you is a URL and (optionally) a description of what to test.

When a web app changes — a button moves, a label changes, a new form field appears — traditional tests break and stay broken until someone manually updates them.

AWT's DevQA Loop re-scans the page after a failure, finds the updated element, and patches the test step automatically. You don't have to touch the test files.

Cloud (no install): ai-watch-tester.vercel.app

Local:

pip install aat-devqa
playwright install chromium
aat dashboard     # opens at http://localhost:9500

From source:

git clone https://github.com/ksgisang/AI-Watch-Tester.git
cd AI-Watch-Tester
make dev && aat dashboard

Use aat devqa when starting from scratch. Use aat loop when you want AWT to also fix your application code.

--verbosity — how many steps run:

• detailed (default): all steps including wait/assert/screenshot

• concise: core actions only (navigate, click, type) — faster

--screenshots — how many images are saved:

• all (default): after every step

• before-after: before + after each click/type/navigate (~70% fewer files)

• on-failure: only when a step fails (great for CI/CD)

# Recommended for daily QA
aat run --verbosity=concise --screenshots=before-after scenarios/

# For CI/CD pipelines
aat run --verbosity=concise --screenshots=on-failure scenarios/

Cloud BYOK keys are encrypted at rest (Fernet/AES-128-CBC).

The local CLI is free forever with no limits.

Yes. For local runs, use the --screenshots=on-failure flag to keep output minimal. For cloud, the API accepts a POST request:

curl -X POST https://your-awt-server.com/api/v1/run \
  -H "X-API-Key: awt_your_key" \
  -H "Content-Type: application/json" \
  -d '{"target_url": "https://staging.example.com"}'

See the CI/CD Guide for GitHub Actions and GitLab CI examples.

• All traffic encrypted via HTTPS/TLS

• BYOK API keys: Fernet-encrypted (AES-128-CBC + HMAC-SHA256) at rest

• Screenshots: auto-deleted after 7 days

• Local mode: nothing leaves your machine

• See our Privacy Policy

License

MIT — free for personal and commercial use.