Keep your tests. Add AI Checks.

Upgrade your automation — don't rewrite it

The delta is three lines. Everything else is your existing test.

def test_checkout(page):
    page.goto("/cart")
    page.click("#checkout")
    page.fill("#card", "4242...")
    page.click("#pay")
    expect(page).to_have_url("/success")

def test_checkout(page, testersai):
    page.goto("/cart")
    testersai.screenshot()         # check cart UI
    page.click("#checkout")
    page.fill("#card", "4242...")
    page.click("#pay")
    testersai.screenshot()         # check success page
    testersai.console()            # check for JS errors
    expect(page).to_have_url("/success")

How it works

Five stages. The travelling dot is the live flow on every call.

If the AI call fails, hangs, is rate-limited, or hits a firewall — the SDK gives up in under a second and your test continues. AI checks never block your real tests.

Where & when to put AI checks

Rule of thumb: wherever a human reviewer would pause and look during a manual run — that's where to drop a check. Interesting states deserve eyes.

Why this SDK

Principles the SDK is built on — encoded in tests, not slogans.

What you can send for analysis

Five endpoints, consistent across every language.

Get user feedback — without recruiting

One call generates diverse end-user personas, each rating the page on usability, accessibility, design, content, and visual (1-10 each) plus written comments. Catches the UX issues that don't show up in functional tests.

AI

Aisha Khan 28 AI

Marketing manager · mobile-first

Visual8

Design6

Usability7

Content7

Features6

Compet.7

Emotion6

A11y5

NPS7

"Hero CTA is clear but mobile layout cuts off the right column. I wasn't sure what 'Vibe Testing' meant from the headline alone."

AI

Chen 41 AI

Senior backend engineer · code-first

Visual8

Design8

Usability9

Content8

Features9

Compet.8

Emotion7

A11y6

NPS9

"Code samples look good and the install section is concise. Minor: the curl example would benefit from a copy button."

AI

Akira 19 AI

CS student · screen-reader user

Visual4

Design7

Usability7

Content6

Features7

Compet.6

Emotion5

A11y3

NPS6

"Some images are missing alt text and the testers row scrolls horizontally without a visible scroll affordance. The hero gradient is hard to read."

# Python — same shape in JS / Java / C# / Ruby
from testersai import TestersAI

ta = TestersAI()
r = ta.analyze_personas(
    screenshot=png,
    page_text=html,
    personas=3,
    persona_traits=["mobile-first", "first-time-user"],
    fail_below=5,           # auto-promote < 5 ratings to issues
)

for p in r.raw["personas"]:
    print(p["name"], p["ratings"])

if r.failed:                       # True if any rating < 5
    raise AssertionError(r.issues[0].message)

{
  "personas": [
    {
      "name": "Aisha Khan", "age": 28,
      "image": "https://testers.ai/img/profiles/aisha.jpg",
      "background": "Marketing manager, mobile-first user",
      "ratings": {
        "visual": 8, "design": 6,
        "usability": 7, "content": 7,
        "features": 6, "competitive": 7,
        "emotional": 6, "accessibility": 5,
        "nps": 7
      },
      "comments": "Hero CTA is clear but mobile layout..."
    },
    ...
  ]
}

Every language × framework

Click a cell to jump to a self-contained page — install, snippet, recommendations. Shareable URL.

30-second quick-start

Python + pytest + Playwright. Same shape in every language.

# 1. install — download the bundle from the Downloads page, then:
pip install ./testersai-python-pytest-0.1.0.tar.gz
# (works identically on macOS, Linux, and Windows — Windows 10 1803+ has tar built in)

# 2. configure
# Get a key at https://testers.ai/sdk or click "Get API key" above.
export TESTERSAI_API_KEY=sk_...
# Windows PowerShell:  $env:TESTERSAI_API_KEY = "sk_..."

# 3. use in your test
from playwright_testersai import TestersAIPage

def test_home_page(page, testersai):
    ta = TestersAIPage(page)
    ta.capture_console()
    page.goto("https://example.com")

    shot = ta.analyze_screenshot()
    console = ta.analyze_console()

    # Findings show up in the pytest report automatically.
    # Fail the test only where you want to:
    if shot.failed:
        pytest.fail(shot.issues[0].message)

Ready to add AI checks to your suite?

Full SDK source, examples, and pre-packaged tarballs for every language. Drop it on any webserver and it works — zero build, zero runtime.