// AI-powered testing for no-code apps

Your app works.
Are you sure?

TestPilot learns your app and tests it like a real user — automatically.
Works on any web app — no-code or hand-coded.

Run your first test — free → See how it works

// First test free · No credit card · No API key needed · Just your app's URL

TP
// no signup needed Try the Sandbox Demo Watch TestPilot learn an app, find a real bug, and heal a broken selector — a fully simulated walkthrough. Explore it live →
WORKS ON ANY APP · LOVABLE · BASE44 · BOLT · V0 · BUBBLE (BETA) · MULTI-ROLE TESTING · SECURITY SCANNING · FINDS REAL BUGS · WORKS ON ANY APP · LOVABLE · BASE44 · BOLT · V0 · BUBBLE (BETA) · MULTI-ROLE TESTING · SECURITY SCANNING · FINDS REAL BUGS ·

// process

Three steps.
Zero code.

01 —

Give it your app URL

TestPilot crawls every page, learns every button, form, and flow. Like hiring a QA engineer who reads the whole manual first.

02 —

Describe what to test

Write in plain English. "Book an appointment and verify it saves." "Create a client, add a job, send an invoice." TestPilot figures out the rest.

03 —

Get a full bug report

Every step logged with screenshots. Bugs flagged by severity. Share it with your team or fix it yourself — TestPilot shows you exactly what broke.

// capabilities

What it can do.

🤖

Scenario Testing

TestPilot acts like a real user — clicks, fills forms, navigates, verifies results. No selectors to write. No scripts to maintain. Just describe the scenario.

🔐

Security Testing

Test that User A can't see User B's data. Checks URL manipulation, IDOR, session persistence, role escalation — the vulnerabilities no-code builders miss.

👥

Multi-Role Testing BETA

Run scenarios across multiple users simultaneously. Dispatcher assigns a job. Contractor bids. Admin approves. All in one test run.

💬

Interactive Mode BETA

Chat with your app through TestPilot. "Go to Clientes. Create Ivan Fernandez." Natural language control over a live browser session.

// api key

Connect your
Claude key.

TestPilot uses Claude AI to browse your app and think like a real user. You connect your own Anthropic account — so you only pay for what you actually use, with no hidden costs on top.

A typical test run costs between €0.10 and €0.50 depending on complexity.

💡 Tip: Set a €10 monthly spending limit in Anthropic's dashboard so you never get a surprise bill.
// Get your Claude API key in 4 steps

01
console.anthropic.com → Create a free account with your email
02
Add a payment method Billing → Add card · Start with €5, enough for 20-50 runs
03
Create your API key API Keys → Create Key → Copy it immediately
04
Paste it in TestPilot You're ready to run your first test ▊
38 Steps, zero retries
0 Code required
Any Web app supported
4 Security scan levels

// embeddable widget

One line. A “Test” button
inside your own app.

Paste a single tag and a floating “Test” button appears in your app — so you and your team can run TestPilot on the live page, no dashboard round-trip. It runs on your own Claude key, and your secret key never touches the browser.

<script src="https://testpilotapp.dev/embed.js" data-tp-token="pk_live_…" defer></script>
🔑
Your key never touches the browser Connect your Anthropic key once. It's stored encrypted (AES-256-GCM) on our server — the embed only carries a public token. Revoke anytime.
💳
Bring your own key Every test runs on your key — you pay Anthropic directly. No markup, and Anthropic's API doesn't train on your data.
🤫
Nothing captured silently Your login token is read only if a user opts into “test as logged-in user” — first-party, over HTTPS, for that one run. Never in the URL, referrer, or history.
👁️
We test your app — we don't harvest it We store only what your report needs — the learned page map + run screenshots — scoped to your account. Never sold or shared, and deletable.

// hidden from your end users by default — your app reveals it only for your team · runs are rate-limited to protect your budget

// pricing

Simple, honest pricing.

You bring your own Claude API key. TestPilot never touches your data or charges hidden AI fees. You only pay for what you actually use.

Solo

€19

per month · up to 3 apps

  • Unlimited test runs
  • Up to 3 learned apps
  • Full self-healing loops
  • Full test history
Get started

Pro

€49

per month · up to 10 apps

  • Unlimited test runs
  • Up to 10 learned apps
  • Full self-healing loops
  • Team collaboration
  • Priority queues
Get started

Agency

€100

per month · unlimited apps

  • Unlimited test runs
  • Unlimited apps
  • Full self-healing loops
  • White-label reporting
  • Custom webhooks
Get started

// Every plan includes full self-healing loops, Scenario Testing, Interactive Test, Multi-Role Testing, and Security + Deep Scan. Higher tiers add more apps, team collaboration, and reporting/webhook integrations.
You only pay for runs that actually tested your app — if TestPilot fails, the run is free.

💡 All plans use BYOK (Bring Your Own Key) — connect your Anthropic account. Typical test costs €0.10–€0.50 in Claude API fees, paid directly to Anthropic.

Find your first bug today.

It takes 2 minutes.

Run your first test — free →