Mr Suricate allows you to create code-free automated tests for the web, mobile, and APIs. You navigate through your application, and Suricapture records your actions. Or you can build the scenario block by block in the editor. Soon, you’ll also be able to dictate it to an AI assistant. The result reads like a series of steps, not a script.
Without a single line of code AI-Assisted Capture Web, Mobile, and APIs
A script written in code becomes unreadable to the rest of the team as soon as its author leaves. And as long as the code depends on a single developer, coverage progresses at the pace dictated by sprint priorities—that is, slowly.
The Numbers
Some organizations still deploy untested code to production.
Tricentis, Quality Transformation Report 2026, 2,501 respondents
cite a lack of skills as one of the main obstacles to maintaining quality at scale.
Tricentis, Quality Transformation Report 2026
the annual cost of poor software quality in the United States, including $1,520 billion in technical debt.
CISQ, The Cost of Poor Software Quality in the U.S., 2022
Sources: Tricentis, *Quality Transformation Report 2026* , and CISQ, *The Cost of Poor Software Quality in the U.S.*, 2022.
Three appetizers
Depending on what you have on hand—an existing screen, a specific idea for the user flow, or a written specification—you’ll end up with the same result.
Suricapture opens a dedicated browser, and you go through the process as usual. Clicks, keystrokes, page changes—everything is captured. You end up with a complete scenario to review, not a blank page.
You take the actions from the list and perform them in sequence. This is the approach to take when the feature hasn't been delivered yet, when the test is targeting a boundary case, or when you're starting from a related scenario.
You describe the flow to be tested to your AI assistant in French, just as you would explain it to a colleague. It writes the scenario in your Mr Suricate space. This is the process when you start from a specification or a ticket rather than from a screen.
The MCP connector is currently being rolled out and will be available in the fourth quarter.
On the platform
A single connection workflow, from recording to verification: Suricapture, your team's library, the editor, text mode, and the execution history.
1
You open the flow in your browser and go through it, just like a user would. Suricapture records every action in its panel: page opens, clicks, text entry, waiting. By the end, the scenario is already there, numbered and easy to review. All that’s left is to save it and refine it.
The work happens during filming, not afterward. That’s what determines whether your script will survive the next rewrite.
Suricapture installs on your computer and runs its own browser—never your personal profile. You choose the platform when you set it up: web, Android mobile, or iOS mobile.
While capturing, Suricapture scans the page and displays elements similar to the one you just clicked on. When the choice is unclear, it flags it with a reliability score, and you make the decision on the spot.
Pre-production behind an .htaccess authentication, login credentials: Suricapture opens the window; you enter them once, and the credentials are sent to an encrypted manager instead of being left in plain text in the script.
Screen resolution, language, geolocation, blocked resources: the replay conditions are saved with the scenario. It will be replayed in the same setting in six months.
2
The library tracks what you're covering. Online subscription, back office, native app, accessibility: each scenario has its own status, last successful execution, and open incidents. You can pause a scenario while it's being redesigned and then resume it afterward. Its content remains unchanged.
3
Action groups break down the workflow, variables hold the data, and metabocks encapsulate repetitive elements. A selector can be changed using a drop-down list. There’s nothing to compile and nothing to indent.
4
Text mode outputs the complete tree structure, one line per action, numbered 1, 2-1, 2-1-3. A product manager or auditor can read it without needing an explanation of the tool. This is what ensures that a test remains understandable even when the person who created it is no longer there.
5
Every run leaves a record: scheduled date, actual date, duration, result, and incident. In this scenario, there were 327 runs—one every six hours—lasting between 19 and 26 seconds. When a failure occurs, the incident identifies the step where it happened.
Details about reports and alerts can be found on the Reporting, Incidents, and Evidence page.
What you can test: your web experiences, your native iOS and Android mobile apps, and your APIs. Not thick clients.
The Basic Unit
A block describes a user action. Three fields—no more. That's what allows someone who has never written a test to edit one.
What about the wait? It isn't hidden in a setting. It's a single block: waiting for the element to become visible. A test that explicitly waits does not become unstable at the first slowdown.
Reuse
The connection appears in almost all of your workflows. The same goes for searching for a folder. You write it once in a metablock, then call it from there.
Exit
Your scripts don't just sit around here. The same script is available in three different formats, depending on who will be reading it.
One line per step. That's what goes into a test case document, a coverage matrix, or an acceptance report, and it can be reviewed in a spreadsheet.
The scenario in .feature, Given, When, Then. The structure of teams working with databases—the one that brings the product and tech together in the same document.
Executable code that can be versioned in your repository, which your developers review just like everything else. What you've built without writing code is now code.
Customer Feedback
What convinced me was the no-code approach. Our QA engineer got the hang of the tool in just a few days, working completely on their own, without needing to bring in developers.
As a software publisher, our release cycles are demanding, and quality is non-negotiable. Before Mr Suricate, our testing efforts relied heavily on manual processes, which took a considerable amount of time and slowed down our deliveries.
She was able to build and maintain the test cases covering our critical workflows on her own and run them at regular intervals. The benefits are very tangible: fewer bugs in production, smoother deployments, and a QA engineer who focuses on analysis rather than repetitive execution.
Avoiding just one critical bug is enough to make the solution pay for itself over the course of a year. I recommend Mr Suricate to any software publisher that wants to professionalize its QA without overburdening its teams.
GO FURTHER
Every component of the platform, from scenario creation to verification of its execution.
Is there a term you don't understand? The automated testing glossary defines it.
YOUR QUESTIONS
Creation, AI-powered data capture, relevant profiles: answers to the most frequently asked questions.
No, and that's the point. Not to create a scenario, not to modify it, and not to review it. The blocks describe user actions, not code.
It observes your interactions as you navigate through your app and derives the corresponding sequence of blocks. You're in charge of reviewing and fine-tuning the results.
QA, product, business, and IT teams all work on the same platform, using the same building blocks. This is what allows test creation to extend beyond the scope of developers alone.
Each block has a clear label, and repeated actions are grouped into reusable blocks. A scenario reads like a series of steps, not like a script.
In the product documentation, which remains the definitive reference for options, selectors, and special cases. This page describes the value and how it works, not the configuration.
The selector is a field in the block, not a line buried deep within a file. You correct it where you see it. The self-healing feature fixes part of it on its own. And if the action is shared by multiple scenarios via a metablock, the correction is made only once.
Yes, in three formats. CSV for populating a test suite or a coverage matrix. Gherkin for teams working with databases. Playwright for retrieving executable code from your repository. Building without code doesn’t mean you’re stuck.
Pause
Twenty seconds, a fake app, bugs that pop up randomly. Click on the bugs, not on the green test results.
Twenty seconds—that's all
Bugs come and go. It's up to you to catch them before they do.
0 bugs fixed
A 30-minute demo of your own app. You'll see a scenario take shape in real time, based on your actual user journeys.