Question 1

What is Stunt Double?

Accepted Answer

Stunt Double deploys a cast of AI actors, each with a distinct persona, to navigate your product like real users. They work through journeys, complete tasks and report exactly what they did, backed by screenshots, transcripts and recommendations.

Question 2

How is this different from traditional user testing?

Accepted Answer

It complements it. Your actors give fast, repeatable feedback at any hour, across personas and devices, so you catch the obvious issues before spending real users’ time. You can bring human testers in too, and Stunt Double folds both into one set of findings.

Question 3

Can I validate a design before I build it?

Accepted Answer

Yes. Point an interview at a Figma prototype and your panel works through the design frames, so you can test the direction while it is still cheap to change.

Question 4

How trustworthy are the findings?

Accepted Answer

Every finding is grounded in evidence: the actor’s reasoning, the steps they took, and screenshots of what they saw. You can judge each recommendation on the proof rather than take it on faith.

Question 5

Can I collect feedback from real visitors too?

Accepted Answer

Yes. Add the feedback widget to your own site with a single script tag and visitors can pin a comment on the exact spot they mean. Each comment arrives with a screenshot, the page URL, the pin position, and device details, grouped by page and by the version of the page it was left on. You reply in the thread and move each one through new, reviewed, resolved, or dismissed, and the author is notified when you resolve theirs.

Question 6

Who can use the feedback widget on my site?

Accepted Answer

Only the domains you allow. The widget is identified by a per-project key that sits in your page source, so we treat it as an identifier rather than a secret: the real control is the domain allowlist, which every widget endpoint checks before answering. You can also require sign-in so that only members of your workspace can read or leave comments, which is what most teams use for staging. Workspace admins can narrow the allowed domains for every project at once, or switch the widget off across the whole workspace.

Question 7

Can I get a design review in Slack?

Accepted Answer

Yes. Connect Slack and Stunt Double reviews the links your team already shares: post a link with something like "can I get some feedback on this?" and it opens the page in a real browser, or renders the Figma frame if you have Figma connected, then replies in the thread with a summary and findings ranked by severity. It works in a direct message straight after connecting, and you can turn it on in a channel so posted requests are picked up without anyone mentioning the app. Up to three links from one message are reviewed.

Question 8

What else can the Slack app do?

Accepted Answer

Anything that is not a link plus a request for feedback goes to the agent. Message it directly or @mention it in a channel and it answers questions about the workspace, lists your projects, actors, checklists, and workflows, and starts a checklist or workflow run for you. Each thread is one conversation, so follow-up messages carry the context of what you were just discussing, and runs are attributed to you once you have linked your Slack identity.

Question 9

Which tools does it work with?

Accepted Answer

Linear, Slack, GitHub, Figma, Vercel, and webhooks, plus a Model Context Protocol server so you can run journeys straight from Claude, ChatGPT, or any AI assistant.

Question 10

How do I get started?

Accepted Answer

Create your first actor in minutes and run your first journey the same day, then upgrade when you are ready to bring in the whole team.

Question 11

How does Stunt Double prevent costly mistakes?

Accepted Answer

Stunt Double puts your product in front of users before you commit resources to building it out. By simulating real user behavior early, you catch usability issues, confusing flows, and friction points while they are still cheap to change, rather than after launch. It's like having a dress rehearsal for your product.

Question 12

What makes this faster than traditional research?

Accepted Answer

Traditional research requires recruiting, scheduling, conducting interviews, and analysis, taking weeks or months. Stunt Double runs AI actors in parallel immediately. You get insights in minutes, so you can iterate while momentum is high.

Question 13

How accurate are AI actors compared to real users?

Accepted Answer

We do not quote an accuracy percentage, because we have not run the controlled study it would take to stand one up. What we can tell you is where actors are strong: exploring flows quickly, across personas, devices and locales, and showing the reasoning and screenshots behind every finding so you can judge each one on the evidence. They are not a replacement for talking to real people. Most teams use Stunt Double for rapid exploration, then validate the critical findings with targeted user research.

Question 14

Can I cancel or change my plan anytime?

Accepted Answer

Yes, you can upgrade, downgrade, or cancel your subscription at any time. Changes take effect at the start of your next billing cycle, and we'll prorate any upgrades.

Question 15

Do you offer startup, education, or non-profit pricing?

Accepted Answer

Yes. Talk to us about our startup, education and non-profit programs.

Question 16

Can AI actors interact with login-protected pages?

Accepted Answer

Yes, you can provide credentials or session tokens for AI actors to access authenticated areas of your product. All credentials are encrypted and can be configured with specific permissions.

Question 17

Can I integrate Stunt Double with other tools?

Accepted Answer

Yes, we offer integrations with popular tools including Figma, Linear, Slack, and more. Enterprise plans can request custom integrations and API access.

Question 18

Is my data secure?

Accepted Answer

Yes. Security and privacy are foundational to how Stunt Double is built. See our security page for the full picture, including encryption, compliance, and data handling.

Support

Browse by topic

Documentation

Guides

Actors & checklists

Feedback widget

Slack

Integrations

Billing & plans

Frequently asked questions

Still need help?