Launch likeyou’ve done it before.

Stunt Double’s actors drive a real browser through your product, prototype or Figma design, and show you authentic insights, backed by evidence.

Start free See a real run

No credit card required.

01 / 08

Every kind of user, on demand.

Powerful perspectives
Each actor brings its own goals, context and habits, so you see your product through their eyes.
Replay, reset, repeat
Actors start every run from a clean slate, so you can rerun the same flow after each fix and know it worked.
Authentic experiences
Actors drive a real browser and stop where a person would, so what you learn comes from real behavior, not guesses.

v1.28What’s new in v1.28Usability scores for interviews, one place for integrations and guided checklistsLatest release

02 / 08

Fix it, then prove the fix.

A checklist run logs every page the actor opened and every action it took, with its reasoning. Fix what failed, rerun the same checklist, and see it pass.

03 / 08

Research without the recruiting.

One guide, a whole panel of actors at once. Follow each one as it works and read each answer at the step it gave it, so a round of research is back in minutes.

Read the synthetic research page

04 / 08

Trust every finding.

Open any finding for the step, the screenshot and the reasoning behind every pass and fail. This one points at the fix: show the refund window with the deposit, before guest details.

The run's journey: 4m 41s end to end, 20 browser steps and one email, then each page the first-time diner opened and each action, with its reasoning beside it. — Friday booking flow, journeyInterview round, liveFriday booking flow, run 03

02 / 08

Fix it, then prove the fix.

A checklist run logs every page the actor opened and every action it took, with its reasoning. Fix what failed, rerun the same checklist, and see it pass.

03 / 08

Research without the recruiting.

One guide, a whole panel of actors at once. Follow each one as it works and read each answer at the step it gave it, so a round of research is back in minutes.

Read the synthetic research page

Interview round, live

Reservation flow: search to confirmationLive

01 Finding the right table02 Completing the booking03 After confirmation

First-time diner

Regular who knows the shortcuts

Screen-reader guest

Guest booking in another currency / locale

Sceptical venue manager

Screen-reader guestThe selected time was not announced.
Guest booking in another currency / localeThe amount needs an AUD label before payment.
First-time dinerThe deposit only became clear at the last step.

04 / 08

Trust every finding.

Open any finding for the step, the screenshot and the reasoning behind every pass and fail. This one points at the fix: show the refund window with the deposit, before guest details.

The run report for the Friday booking flow, run 03: a weighted score of 66.7%, one check failed and two passed. The failed check, deposit amount and refund window are stated before payment, is open on its reasoning and the payment step it was judged on. — Friday booking flow, run 03

05 / 08Run 03

Your intuition, backed by evidence.

Fig. 01
Recording4m 48s
Screenshots and recordings: every screen the actor saw, and the run on tape.
Fig. 02
JourneyFirst-time diner
User journey: each step the actor took, with the reasoning behind it.
Fig. 03
Map5 screens
Topology: every screen the run passed through, on one map of the product.
Fig. 04
HistoryRun 01 to 03
Run history: what changed since the last run, check by check.

06 / 08

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.

What Runner can do

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

07 / 08

Works where you work.

Start runs from the tools you already use. Findings land in the same place.

Claude and any MCP client
Run checklists and interviews from a conversation, without leaving the conversation.
Slack
Ask for a run in a channel and get the findings back where your team already talks.
Linear
Delegate an issue to Stunt Double and get the answer, or the run, back in the issue.
GitHub
See what a pull request broke before it merges, right on the PR.
Figma
Have actors review a frame in character, with what they find pinned to the canvas.
Vercel
Point actors at your latest preview or production deployment, even behind deployment protection.

See all integrations

08 / 08

User-centered design for the age of AI.

Your next user may not be human.
Agents now browse, compare and buy for real people. Send an actor to see where they get stuck.
Product insights for agentic use cases.
Run the same checklists and interviews for agents as for people, and fix what blocks them first.

websites scored by AI agents: 550¹
their average score, out of 100: 76.7¹
score Poor or Critical: an agent struggles to finish: 6%¹
scoring runs behind the ranking: 3,542¹

1The Stunt Double Index (index.stuntdouble.io), listed websites with a score, as of October 13, 2026. Poor and Critical are the Index's bands under 58.

Browse the Index

Ship with the proof in hand.

Point an actor at your product, a preview or a prototype. Your first run is back in minutes, with the screen and the reasoning behind every finding.

Talk to the founder Start free

No credit card required.

Launch likeyou’ve done it before.

Every kind of user, on demand.

Powerful perspectives

Replay, reset, repeat

Authentic experiences

Fix it, then prove the fix.

Research without the recruiting.

Trust every finding.

Fix it, then prove the fix.

Research without the recruiting.

Trust every finding.

Your intuition, backed by evidence.

Set it all up in one sentence.

Works where you work.

Claude and any MCP client

Slack

Linear

GitHub

Figma

Vercel

User-centered design for the age of AI.

Your next user may not be human.

Product insights for agentic use cases.

Ship with the proof in hand.

Launch likeyou’ve done it before.

Every kind of user, on demand.

Powerful perspectives

Replay, reset, repeat

Authentic experiences

Fix it, then prove the fix.

Research without the recruiting.

Trust every finding.

Fix it, then prove the fix.

Research without the recruiting.

Trust every finding.

Your intuition, backed by evidence.

Set it all up in one sentence.

Works where you work.

Claude and any MCP client

Slack

Linear

GitHub

Figma

Vercel

User-centered design for the age of AI.

Your next user may not be human.

Product insights for agentic use cases.

Ship with the proof in hand.