# Stunt Double: product testing and user research, run by AI users

> Launch like you've done it before. Intelligent AI agents act as users using your live product, prototypes or Figma designs, get stuck where your customers would, and report back with insights and recommendations backed by evidence.

[Run your first scene](https://app.stuntdouble.io/login) · [Try the live demo](https://app.stuntdouble.io/demo) (free, no sign-up) · [Pricing](https://www.stuntdouble.io/pricing)

## Know where users get stuck, before and after you ship

Point Stunt Double at a question about your product and get an answer you can act on today.

- **Know before you build.** Understand frictions with unprecedented observability. Put a prototype, a flow or a headline in front of actors in minutes, long before you commit engineering time to it.
- **Catch problems before your users do.** Every change is checked against your requirements automatically, with the screenshot and the plain-language reasoning behind each pass and fail.
- **Ship changes that move the metric.** Optimise toward the outcome you actually care about, then see exactly what moved it, so you iterate on evidence instead of opinion.
- **Test from where your users are.** A cast of users browses from the regions, devices and locales you serve, surfacing the friction you would never hit on your own setup.

More detail: [Product](https://www.stuntdouble.io/product).

## A stand-in for your users at every stage

Interviews before you build, checks on every change, optimisation once it is live: three stages, one cast.

### 1. Discovery: hear it in their words

Run guided, qualitative interviews with AI personas to explore needs, reactions and assumptions early, long before you commit to a direction. Follow every step live, then read the synthesised themes.

### 2. Validation: check it against your requirements

Turn your requirements into checklists that actors run automatically on every change, with the screenshots and the reasoning behind every result so you can trust each pass and fail. Covers automated checklists, usability and accessibility, and regression sweeps on every deploy.

### 3. Improvement: move the metric that matters

Optimise toward business goals with behavioural psychology and cognitive bias, then iterate on the changes that actually move the metric rather than the ones that simply feel better. Covers conversion optimisation and A/B iteration round over round.

## Your next user is an agent

Some of what goes through your signup, checkout or booking flow is already an agent acting for a person: a shopping assistant, a browser agent, an automation filling the form. Nobody is watching it try.

- **Cast an agent as the user.** An actor is an agent driving a real browser. Give it the task an assistant would be given, and watch it attempt your flow.
- **Observable orchestration.** Every run is a sequence you can read, with a screenshot at each step. Put a checklist in the deploy loop and it runs on every change.
- **Serve both kinds of user.** The fixes are mostly the same ones that help a screen-reader user, so one round finds them for people and agents at once.

More detail: [Synthetic research](https://www.stuntdouble.io/synthetic-research).

## Ask your Runner

Runner is your on-call assistant. Ask in plain language and it casts actors, builds checklists, runs automations and finds answers, then takes you straight to the result.

- **Skip the busywork.** No forms or menus. Say what you want and Runner works out the steps.
- **Gets it done for you.** Creates actors, checklists, interviews and automations, not just answers about them.
- **Understands your context.** Sees the page you are on, so "debug this run" or "fix this check" just works.
- **Plugs into your agent stack.** Ask from Slack, drive a run from Claude, and land the findings in Linear as issues your team can pick up.

## For AI agents

Stunt Double is also available to agents over the Model Context Protocol at `https://app.stuntdouble.io/api/mcp` (Streamable HTTP, OAuth 2.1).

- [llms.txt](https://www.stuntdouble.io/llms.txt): site summary and when to use Stunt Double.
- [Documentation](https://www.stuntdouble.io/support/docs): concepts, tutorials and the API.
- [MCP server docs](https://www.stuntdouble.io/support/docs/api/mcp): connect an agent.
- [MCP manifest](https://www.stuntdouble.io/.well-known/mcp.json): transport, auth and tool list.
- [OpenAPI](https://www.stuntdouble.io/openapi.json): public HTTP endpoints.
- [Sitemap](https://www.stuntdouble.io/sitemap.xml): every public page.

## Your cast is ready to start

Create your first actor in minutes: [Run your first scene](https://app.stuntdouble.io/login).