Web · Mobile · API — one studio

The AI test studio that fixes its own failures — privately.

Record tests without code, run them on your own machine, and let QA Pilot — your AI test agent — find the cause of a failure, fix it and re-run it. Bring any AI model, or run fully offline.

Book a demo Take the product tour ▸

Runs on your machine
Any AI model
No per-run cloud fees

QA Pilot · Agent mode

Fix my latest failure

FAILED Checkout › Step 7 · Click “Place order” — element not found (30 s)

ANALYSE Compared recorded element with current page and failure screenshot

CAUSE Locator — button id changed #place-order → #submit-order

CHANGE Update step locator · asks approval · can be undone

RE-RUN Checkout › running 12 steps…

PASSED 12/12 steps · fix remembered for similar failures

Works with the AI you already trust

Google Gemini
OpenAI
Anthropic Claude
Azure OpenAI
OpenRouter
Groq
Mistral
Ollama (offline)
LM Studio
Google Gemini
OpenAI
Anthropic Claude
Azure OpenAI
OpenRouter
Groq
Mistral
Ollama (offline)
LM Studio

The problem

Test automation still fails people quietly.

Recordings vanish. A 50-step flow saves with zero steps, and nobody is told.

Tests pass without testing. A click lands on the wrong layer and the step never really happens.

One journey, four tools. Web, mobile, API and database checks live in separate products.

Accessibility comes last. It is audited after release, by someone else.

Results stay in the tool. They never reach the board, the channel or the pipeline.

Why AI Automation Studio

Built for trust, not just for demos

Local-first & private

Tests, screenshots and history stay on your machine. Typed passwords and secrets are never sent to any AI model.

Bring your own AI

Gemini, OpenAI, Claude, Azure OpenAI, any OpenAI-compatible service — or a local model with zero internet.

One test, every layer

UI, API, mobile, database, email and accessibility are steps in the same flow.

Honest failures, AI fixes

A step that didn’t really happen fails the run. QA Pilot then finds the cause, fixes it and re-runs.

Every action, one glance

Ten building blocks, one tap away

Go beyond clicks and typing — every block is a normal step you can review, edit and replay.

Add API

Mobile steps

Custom JavaScript

Extract value

Element presence

Conditional flow

Loops

Database query

Open / validate email

Accessibility scan

Product tour

See the studio, screen by screen

Swipe through or click a slide to view it full size.

Introducing AI Automation Studio — tests dashboard

The problem: test automation still fails people quietly

QA Pilot designs tests, fixes flaky ones, explains failures

Suites, schedules and a full build history

Runtime engine: same tests, any pipeline

Azure DevOps, Jira, Slack, GitHub and TestRail integrations

Download the product overview (PDF)

The studio

Everything a QA team needs, in one app

Record without code

Click through your app; every action becomes a readable step with smart locators.

AI steps

Describe a step in plain English and let your chosen AI model perform and validate it.

AI self-heal

When an element moves or is renamed, the AI finds the same control and the run continues — clearly marked in results.

Suites & schedules

Group tests, run them on a schedule with test data, browser options and per-test video.

Accessibility scans

Scan a whole page or part of a page; issues show up in every build.

Full build history

Every build with its duration, results, screenshots and accessibility issues.

Android & iOS recorder

Emulators, simulators and real devices — local or remote. Click to tap, drag to swipe, live preview.

Smart waiting

No fixed pauses — each step waits for its element and acts the moment it appears.

Fallback locators

If a locator misses, saved alternatives are tried. Accept the one that worked with a click.

Shared objects

Reuse elements across tests; change a locator once, fix every test.

Runs & reports

Per-step screenshots, timings and JUnit XML for your CI.

Web + phone in one test

Hand steps to a phone mid-test in plain English — the AI reads the live screen on every run.

REST, SOAP & GraphQL

Collections, every common auth type including OAuth 2.0, Digest, NTLM and AWS Signature.

Verification scripts

JavaScript checks with familiar keywords and Postman-style pm.* API.

Schema validation

JSON schema, XSD, GraphQL schema and OpenAPI / Swagger.

AI API testing

Point it at an OpenAPI spec: positive, negative, edge-case and security tests with a compliance report.

Data-driven suites

Profiles, encrypted secrets, data files and retry of failed runs.

Import everything

Postman, OpenAPI, WSDL, SoapUI, cURL and whole Katalon projects.

Real email & OTP testing

Use your own Gmail inbox. Fresh address per run, read the email, extract the OTP — no paid test inboxes.

Integrations

Raise bugs with screenshots in Jira, Azure DevOps and GitHub; Slack build results; TestRail and Azure Test Plans sync.

MCP servers

Connect any MCP server to QA Pilot — with secure sign-in and per-tool approval.

Analytics

Local dashboards for pass rates, failures and run history.

Email reports

Send build results to your team automatically.

Database checks

Run a query mid-test and assert on the data your UI just created.

QA Pilot · AI that stays editable

Creates, runs, schedules and fixes your tests.

An AI test agent that knows your tests, suites, runs and failures — across web, mobile and API. Ask in plain English; it does the work and shows every step.

Create test cases from a feature description or requirement, then record them as real, editable steps.
Run them on Chrome, Edge or Firefox, on Android and iOS emulators, simulators or real phones — one test or a whole suite.
Schedule suites nightly or on every build, with test data, browser options and video per test.
Fix my latest failure — finds the cause (locator, timing, wrong screen, data, environment or app bug), makes the smallest fix and re-runs, up to three attempts.
Test an app on a real phone, write automation code, or analyse a failed CI build.
You stay in control — every change asks for approval and can be undone. Real app bugs are reported, not hidden.
Learns from fixes — fixes that pass are remembered and tried first next time.

3platforms in one test: web, mobile, API

10building blocks beyond clicks and typing

1engine from the desktop to CI

7+AI providers, incl. fully offline

One conversation · whole test lifecycle

Watch QA Pilot take a test from idea to nightly run

QA Pilot

shop.example.com/checkout

Card number

Pay now

Web · Chrome

4829

Confirm

✓ Order placed

Android · Pixel 8

aiasc · Runtime Engine

$ aiasc -runMode=console \
  -testSuiteCollectionPath="Test Suites/Collections/Nightly" \
  -executionProfile=staging -retry=1 \
  -retryStrategy=failedExecutions -report.folderPath=./Reports
INFO  AI Automation Studio Runtime Engine 1.1.0
INFO  START Web/Login suite · Chrome (headless)
INFO  API/Employee API contract suite — PASSED
INFO  Mobile/Checkout — PASSED
Total: 26 | Passed: 25 | Failed: 1 | Time: 4m 12s · Exit code: 1

Runtime Engine

Same tests. Any pipeline.

The aiasc engine runs web, API and mobile suites from the console — in parallel, with retries, profiles, reports and CI-ready exit codes.

GitHub Actions, Jenkins, Azure Pipelines — or Docker.
Runner app — pick suites, choose headless or a browser, press Run.
Retries & profiles — rerun only failed executions, switch environments with one flag.

Security & privacy

Built for teams who can’t send their data to someone else’s cloud

Banks, healthcare, government and enterprise teams choose local-first testing.

Local storage

Tests, recordings, videos and reports are stored on your machine — not uploaded.

Secrets stay secret

Passwords, tokens and masked variables are hidden in logs and reports and never sent to the AI.

Offline AI option

Run a local model through Ollama or LM Studio for fully air-gapped AI.

Encrypted credentials

Sign-in tokens are encrypted with a key held in the system keychain.

Approval for every change

QA Pilot asks before it changes anything, and every change can be undone.

Your own accounts

Integrations use your Jira, GitHub, Slack and Gmail accounts directly.

Compare

How we’re different

	AI Automation Studio	Cloud low-code tools	Legacy test suites	Open-source frameworks
No-code recording	✓	✓	✓	—
Web + mobile + API in one tool	✓	Partial	✓	Separate tools
Web and phone steps in the same test	✓	—	—	Custom code
Database & email steps built in	✓	Partial	Partial	Custom code
Data stays on your machine	✓	—	Partial	✓
Choose any AI model / offline AI	✓	—	—	—
AI agent that fixes and re-runs failures	✓	Locator healing only	Locator healing only	—
MCP server support	✓	—	—	—
Console engine for CI with exit codes	✓	Cloud only	✓	✓
No per-run or parallel-run fees	✓	—	Partial	✓

Switching is easy

Import Katalon projects, Postman collections, OpenAPI, WSDL, SoapUI and cURL. Groovy scripts convert to JavaScript automatically.

Plan your migration

Results where teams work

Plugged into your workflow

Azure DevOps Jira GitHub Slack TestRail Azure Test Plans MCP servers Email reports

Banking & fintech

Sign-up with email OTP, KYC on mobile, balance checks via API — with data kept in-house.

E-commerce

Checkout across web and app, payment APIs, and self-healing when the storefront changes.

Healthcare

Test patient portals without uploading sensitive screenshots or data to a third party.

QA agencies

One studio for every client’s web, mobile and API projects, with each client’s own AI keys.

Pricing

Simple licences. No per-run fees.

Use your own AI key — you only pay your AI provider for what you use.

Starter

For individual testers

Web, mobile & API testing
Bring your own AI
QA Pilot (Ask mode)
Email support

Get a quote

Team

For QA teams

Everything in Starter
QA Pilot agent: fix & re-run failures
Runtime Engine for CI
Integrations: Jira, Azure DevOps, GitHub, Slack
Priority support

Book a demo

Enterprise

For regulated organisations

Custom

Everything in Team
Offline / private AI setup
Onboarding & migration help
Custom integrations
SLA

Talk to us

FAQ

Questions, answered

Do I need to write code?

No. Record tests by clicking through your app. Engineers can still add JavaScript checks and custom steps when they want to.

Which AI models can I use?

Google Gemini, OpenAI, Anthropic Claude, Azure OpenAI and any OpenAI-compatible service such as OpenRouter, Groq or Mistral — or a local model through Ollama or LM Studio.

Does my test data leave my machine?

No. Tests, videos, reports and credentials are stored locally. Only the context needed for an AI step is sent to the AI provider you choose — and never typed passwords or secrets.

Will QA Pilot change my tests without asking?

No. Every change asks for approval first and can be undone. If the failure is a real app bug, it reports it instead of “fixing” the test.

Can I run tests in CI?

Yes. The Runtime Engine (aiasc) runs the same suites from the console, in parallel, with retries, profiles, reports and CI-ready exit codes — in any pipeline or in Docker.

Which platforms are supported?

Web browsers, Android and iOS (emulators, simulators and real devices), and REST, SOAP and GraphQL APIs.

Can I move my existing tests?

Yes. Import Katalon projects, Postman collections, OpenAPI / Swagger, WSDL, SoapUI and cURL.

How do I get access?

Book a demo below or message us on WhatsApp. We’ll walk you through the studio on your own app and set you up with a trial licence.

Book a demo

See it fix a failing test — on your own app.

A 30-minute call: we record a flow from your product, break it, and watch QA Pilot fix it. Then you get a trial licence.

Live demo on your web, mobile or API flow
Migration plan from your current tool
Trial licence for your team

Chat on WhatsApp