Skip to main content

Product

Overview
Features
Synthetic research
Topology
Use cases
Integrations
Pricing
Changelog
Download for Mac

Learn

Documentation
Guides
Blog
Glossary
Alternatives
All resources
Stunt Double Index (opens in new tab)

Help

Help center
Contact
Feedback
Status (opens in new tab)

Company

Design
Brand
Investors
Security
Privacy
Terms
MCP server (opens in new tab)
llms.txt

© 2026 Turnout Labs Ltd. 66 Paul Street, London EC2A 4NA. Company No. 15708750.

“Stunt Double”, “Turnout Labs” and “Aictors” are registered trademarks of Turnout Labs Ltd.

A stand-in for your users, at every stage.

Actors work through your product in a real browser and bring back the screens, steps and quotes behind what they did. You see where people get stuck before you ship.

Run your first scene See a real run

The Parlour Web project in Stunt Double: needs attention, ranked worst first (the Friday booking flow failing at 66.7%, a new comment to triage, two open issues), Ask Runner, and the checklists with their last results. — Parlour Web, project overviewRanked worst first

The Parlour Web project in Stunt Double: needs attention, ranked worst first (the Friday booking flow failing at 66.7%, a new comment to triage, two open issues), Ask Runner, and the checklists with their last results. — Parlour Web, project overviewRanked worst first

Behavior, not opinions.

Ask a simulated user about your page and it says it loves it. Send an actor through the same page and it shows you where it got stuck.

Film00:38 · 24 Sep 2026

Check every change before it ships.

Write the flows that matter as plain-language checks, and actors run them on every preview deploy. Each failure comes back with the step, the screenshot and the reasoning.

Plain language: Describe each check in a sentence and an actor runs it in a hosted browser.
Evidence on every verdict: Every check comes back passed or failed, with a screenshot and the reasoning.
Replays: After its first run a checklist replays the recorded path, so repeat runs are faster and land the same way.

A checklist run for the Friday booking flow: a weighted score of 66.7%, one check failed and two passed, the failed check open on the payment step it was judged on. — Friday booking flow, run 03

A checklist run for the Friday booking flow: a weighted score of 66.7%, one check failed and two passed, the failed check open on the payment step it was judged on. — Friday booking flow, run 03

Hear from a panel this afternoon.

Write a guide, cast a panel, and each actor works through your product while it answers. The round comes back as themes, each citing the quotes and screenshots behind it.

Side by side: Participants run at the same time, so a five-person round takes an afternoon.
Themes with receipts: Each theme cites the sessions, quotes and screenshots behind it.

How a study works

An interview round, Baseline guest journey: five of five participants completed, three findings, four recommendations, and a summary of the confidence gaps they ran into. — Baseline guest journey, round report

An interview round, Baseline guest journey: five of five participants completed, three findings, four recommendations, and a summary of the confidence gaps they ran into. — Baseline guest journey, round report

Users from where your users are.

Each actor browses from a city you pick, on a phone, tablet or desktop. It has its own inbox, so sign-up codes and confirmation emails arrive and get used.

An inbox of its own: Verification emails, magic links and one-time codes reach the actor mid-run.
Ask it afterwards: Chat with an actor about a run, and it answers from what it did.

The first-time diner's inbox at stdbl.com: three booking confirmations from Ondine, the newest for 4 on Friday 16 October at 7:15 pm, reference OND-49094. — First-time diner, inbox

The first-time diner's inbox at stdbl.com: three booking confirmations from Ondine, the newest for 4 on Friday 16 October at 7:15 pm, reference OND-49094. — First-time diner, inbox

Test it at every stage.

Anything with a URL gets the same run, so you can start while a change is still cheap.

Prototype: Point actors at a Figma Make site, a preview deployment or a Claude artifact, and each reviews it in character.
Pull request: Checklists run on each preview deploy and report on the pull request that changed the flow.
Production: Automations re-run checklists on a schedule or an event, so a check that starts failing shows up before a complaint.

Testing designs

The Friday booking flow checklist and its runs: the latest completed by an agent, 66.7% and failed, in 4m 48s. — Friday booking flow, runs

The Friday booking flow checklist and its runs: the latest completed by an agent, 66.7% and failed, in 4m 48s. — Friday booking flow, runs

Set it all up in one sentence.

Describe what you need. Runner sets up the work and takes you straight to the result.

Skip the blank page. Runner writes the discussion guide for your interview.
Cover a flow in a sentence. Runner writes the checklist from what you describe.
Stop rerunning by hand. Runner schedules the repeat work, or runs it after every deploy.

Runner, asked to set up a checklist on Parlour Web for a first-time diner booking a table for four next Friday at 7 pm, creates the checklist Friday booking flow with three checks, an automation that re-runs it after every deploy, and starts a run as the first-time diner.

Runner

Set up a checklist on Parlour Web: a first-time diner books a table for four next Friday at 7 pm. Run it now and after every deploy.

I’ll write the checks from what a first visit needs to see.

Created checklistFriday booking flow

Party size is visible before choosing a time slot.
Deposit amount and refund window are stated before payment.
Confirmation shows date, time, party size, and venue address.

Created automationAfter every deploy

The first-time diner is booking now, and the checklist runs again after every deploy. Open the run

Ask anything. / for commands

Findings land where your team works.

Every run reports into the tools you already use. A failed check becomes an issue with its evidence attached.

See every integration

Linear. Failures arrive as issues, ready to triage.
Slack. Run summaries and alerts in your channels.
GitHub. Results on the pull request that changed the flow.
MCP. Claude and other agents run your workspace.

The issue the run raised: Deposit refund window is easy to miss, open, high severity, with the fix in its detail and Michael Parker assigned. — Issue raised by run 03High

The issue the run raised: Deposit refund window is easy to miss, open, high severity, with the fix in its detail and Michael Parker assigned. — Issue raised by run 03High

Know where your users get stuck, before they do.

Run your first scene