AI Testing Agent

Your AI QA engineer.
It writes the tests, runs them, and fixes them when they break.

Point ReTibe at your codebase. The agent reads it, writes real scenarios, runs them on a schedule in real browsers, catches visual regressions and security holes, and files the bug before anyone opens a ticket.

Start free
Or talk to the agent yourself live demo below

Free plan forever: 20 web tests and 20 bot tests a month. No card.

ReTibe IDE — checkout.json Agent active
You
Execution log idle
waiting for the run to finish
67 s
median web test, measured across real runs
17
security check categories, 11 of them on by default
3
engines: Chromium, Firefox, WebKit
48
engine actions, extracted from the code itself
The honest part

You already run Playwright through MCP. It is free, it is familiar, and it is not a test suite.

Playwright MCP drives a live browser, which is exactly right for looking around: open the page, click things, see what is there. Then the session ends and nothing remains. No saved scenario, no report, no history, no schedule. Next week you explain the same flow to the model again.

After the session
Playwright MCP
ReTibe MCP
What is left
Nothing
A saved scenario, a report, run history
Running it again
Explain the whole flow again
One call by id
On a schedule
Only while you are sitting there
Cron, daily, weekly, email on failure
Catching a silent UI break
Only if the model happens to look
Pixel diff against an approved baseline
A test that passes and checks nothing
Nobody notices
Validator flags it before a browser opens

Keep Playwright. Really.

They are not competitors, and we say so in our own docs. Scout the page with Playwright MCP, then hand the flow to ReTibe so it keeps running after you close the laptop.

Free is not the hard part

Our free plan runs 20 web tests a month, forever. The cost of the habit is not money, it is the regression nobody was watching for.

One command to try it

Add the MCP server next to the one you already have and ask the same agent to pin one flow.

See the connector

Guide for the agent: English · русский

How it works

Point it at the code. Walk away.

STEP 01

It reads your project

A progressive L1 to L4 scan pulls routes, forms, auth flows and real selectors into a 300K character context. No guessing at markup it has never seen.

STEP 02

It writes and validates

10 to 50 scenarios in minutes, each checked by a validator that catches invented actions and assertions that would pass on anything.

STEP 03

It keeps running them

Nightly on a cron, in three engines, with pixel diffs, a security score, an emailed report and a Jira bug when something fails.

Visual regression

Baselines, pixel diffs, an approve and reject queue, and a threshold you control.

Security Scan

TLS, headers, XSS, SQLi, CORS, clickjacking, traversal. Scored, with proof and a fix.

Flakiness detection

Stability heatmap per step and a 0 to 100 flakiness score, so you know which test is lying.

Testing calendar

Cron, daily, weekly or monthly, with colour-coded history and email alerts.

CI and Jira

GitHub Actions, GitLab CI, JUnit XML for Jenkins, and Jira bugs created on failure.

Figma comparison

Diff the live page against exported frames and hold the build to the design.

Run history with timeline chart and duration trend
Run historyTimeline, duration trend, flaky detection
Testing calendar week view with scheduled runs
Testing calendarSchedules, statuses and email alerts
Detailed HTML report with screenshots per step
ReportsEvery step, screenshot and timestamp
Workspace

More than a test runner

Test organisation, run history and scenario recording, in one workspace.

Timeline chart

Every run on one timeline, with duration and success trends.

Stability heatmap

Step-level heatmap, so you see instantly which step fails most often.

Flakiness detection

Unstable tests found automatically, with a flakiness score from 0 to 100.

Failure analysis

Root cause of a failure, with recommendations and a run-to-run comparison.

Key metrics
Success Rateshare of passing runs Avg Durationaverage length Min / Maxfastest and slowest run Flakiness Scorestability rating Trendimproving, stable, degrading ExportPDF report, scenario JSON
retibe.com / analytics
Run history dashboard with totals, trend by day, distribution by type and test duration charts

Organise the suite

Checklists with pass and fail tracking and coverage stats
Parameters for credentials, URLs and selectors, stored once
Fixtures for setup, teardown, login and API calls
Page Objects so a selector change is one edit
Environments for dev, staging and production, isolated
Checklists workspace with progress tracking
ChecklistsStructured testing with progress and statistics
Pricing

Cheaper than the release you rolled back

Free

Enough to see it work on your site

₽0forever
Start free
20 web and 20 bot tests a month
2 security scans, 1 baseline
Calendar, AI assistant, MCP access
Corporate

For a QA function with SLAs

4 999/ month
Talk to sales
Unlimited tests, scans and templates
10 team members, white label
Priority support and webhooks
Compare every limit
Live demo · MCP

Ask the agent. Watch it work.

Claude connected to ReTibe over MCP, right here on the page. Pick one of the prompts that work, or write your own, and see the tool calls, the validator and the run result.

claude — retibe mcp demo mode

Prompts that work

Write a ReTibe scenario: open https://example.com, check the heading exists, go to /pricing, confirm the form is visible. Validate it, run it and show me the result.
Run webtest-123 failed. Work out why and propose a fix.
Check my scenarios for assertions that do not actually assert anything.

Prompts that do not

Test my site.

Too vague. The model does not see your page and will not guess what matters. Give it a URL and say what must be true after each step.

Liked what it did? The same connector runs in your own editor. Issue a key, add one MCP server, and the agent you already use can write and keep your tests.

Get your API key Read the setup guide
Try it right now

Website testing

Automatic website and app testing in real time

Test settings

A demo of the console. Free accounts get 20 real runs a month.

Execution log

ready
Ready to run...

Test preview

https://example.com
Screenshot will appear when the test runs
Security score appears with the report
FAQ

What people ask first

Do I have to drop Playwright?

No. Keep it for exploration. ReTibe is where a flow goes once you want it checked on every release without you being there.

Will the agent write tests that pass and check nothing?

That is the failure mode we built the validator for. It runs before a browser opens, flags assertions that cannot fail, and there is a /harden prompt that hunts them across your whole suite.

Do I need to write code?

No. Scenarios are JSON, and you can record them in the browser, describe them to the assistant, or let the agent generate them from your codebase.

How long does a run take?

Median web test is 67 seconds across real platform history, with the slow tail around eleven minutes. Runs are asynchronous, so nothing blocks while it works.

Hire the QA engineer that never sleeps

Twenty runs a month, free forever. Enough to point it at one real flow tonight and read the report in the morning.