Changelog

Three check types: pass/fail blocks the run, weighted feeds a score judged against your threshold, informational records what the actor saw and is never judged. A weighted check can carry a negative weight, deducted when it fails and never added to the maximum, so a rubric written "out of 80" stays out of 80. Name the ranges a score can land in and a run reports its grade beside its score. A check can run only when the one before it passed, or only when it failed; an unmet condition is skipped rather than scored, and checks reorder from the flow editor. Existing checklists keep the old all or nothing rule.

Every run is read against the last completed one, so the report names what changed. Informational checks compare as text, which is what catches $29 per month becoming $39. Each check carries its own evidence, from the tab the run is actually on, shown as the whole frame rather than an 80 pixel strip.

Pins, replies and resolutions land in the Comments view with no reload, and the detail dialog updates while it is open. Comments also get a section on the project overview. A widget comment links to a Stunt Double account from a signed-in session or a matching member address, so that author can resolve their own thread from the page.

When the page is rebuilt, older comments resolve on their own. You do not have to close them by hand.

Generate prompt writes a brief for one comment, its thread and where it sits on screen, ready to paste into your agent.

Admins cap overage for the billing period and set up to five alert thresholds. A cap either pauses runs or only alerts, and every run gate honours it.

The address step is optional, and a fourth type, clickable prototype, covers a Figma Make link, a v0 or Claude artifact or a preview deployment.

summarise_feedback returns the dashboard’s themes, questions and brief, and list_feedback carries the page, element, version and replies. The checklist tools and Runner now read and write check types, weights, conditions and the pass threshold.

The widget outlines the control under the cursor, pins the comment to it and crops the screenshot around it, so a digest names the Billing settings screen and the Change plan button rather than a route fragment. Hash-routed apps show feedback per screen.

Every comment is stamped with the page version it was left on, and the review screen numbers them into a timeline you can filter. Declare a version with data-sd-version, then choose what a release does to older comments.

Type while an answer streams and it queues, or send it now to steer the run in progress. Up arrow edits and re-runs an earlier message, /compact summarises a long conversation, and images paste into the composer.

A review opens menus and follows the flow as the project’s default actor, carrying that actor’s sign-in, tools and knowledge. An ambiguous request gets one question back in the Slack thread.

Fill the Guidelines library from a published Storybook or a GitHub repository, so rules cite the components, variants and tokens you actually ship. Apply rules to design reviews in bulk.

A synthesised round fits two to five minutes, with narration written to a length budget. The finished render is reachable from the interview itself.

A device worker recovers on its own once a dropped heartbeat resumes, so closing a laptop lid no longer leaves a dead entry in the list.

Up to five images in the Feedback dialog, and Cmd/Ctrl + Enter to send. The command bar ranks results as you type, inbox items deep link to the run they are about, and the Mac app renders release notes as text.

A participant that times out is cancelled and reconciled rather than left in progress. Scheduled work wakes when it is due instead of polling, and an error is ticketed with its context and trace.

Watch what your participants did instead of reading about it: the screen each one used and the question framing each moment. It is a synthesised edit rather than a screen recording, so a round plays at the length of a findings readout. Single-participant films are there on demand.

Every theme with video evidence carries a Watch control that seeks to the moment behind it, so you can show a stakeholder the proof in one click. Shared report emails carry the poster frame and a link back, never the footage.

A Live tab shows which participants are running and what each has reached, so you can react to a round early rather than waiting for the report.

A Transcripts tab puts all of a round’s transcripts one click apart, so you can compare how different people answered the same question instead of opening a participant at a time. A transcript being written right now fills in live.

Keep the artefacts a project is judged against in one place: links, Figma files and uploaded assets. Actors read them on every run, guideline drafting works from them, and they show up in search. Change a resource URL once and everything pinned to the old one follows.

Import a project’s rules from a public web page or a GitHub repository, so findings cite your design system, brand guide or contribution rules rather than a generic standard.

A checklist records its first run as a script and replays it after that, so repeat runs are faster and land the same way every time. It re-records itself when the project’s guidelines change, so the script never drifts from the rules.

One search over projects, actors, checklists, interviews, findings, feedback, goals and resources. Open it from the command bar, or call it from your own assistant over MCP.

Drop a page or a Figma link into a mapped channel or a DM and the review comes back in the thread, findings ordered by severity. Replies now stream in with a live status line, the current channel comes along as context, and setup is one step.

Give actors your issue tracker, feature flags or seed-data API mid-run. Admins register remote MCP servers in Settings, credentials are held as encrypted secrets, and each actor is opted in one at a time.

Actors run on Claude Sonnet 5 at medium reasoning effort, holding run quality steady for fewer thinking tokens, with Opus 5 the default where the heavier model runs. An actor pinned to an older model keeps it.

Signing up now walks through one stepped view that seeds a project and a small cast of actors, so a new workspace has something to run on day one.

Retire a project or a workspace in one step that always completes, and bring it back later. An archived project stops running and leaves every surface including search; an archived workspace releases its name, so you can start again on the same URL rather than living with acme-2.

Actor chat and Runner run on a rebuilt durable architecture. A long browser task keeps going instead of hitting a request limit, refreshing the page rejoins the answer already streaming, and the first reply starts sooner.

Usage stats moved out of settings into a dedicated Metrics view in the sidebar, reachable from the command bar.

Pull to refresh reloads on mobile, first paint is quicker on web and desktop, and long interview recordings save reliably however big they get.

A new Security tab in workspace settings, admin only. Switch the public comment widget, public share links and self-hosted workers on or off for the whole workspace; restrict the domains actors are allowed to browse with an allow list or block list; cap which origins any project may serve the widget to; and decide who can invite members and which email domains they may be invited from. Every control is a ceiling, so it can only narrow what a project or worker already allows, never widen it. Turning public sharing off also revokes share links that already exist.

The create-checklist dialog now opens on a library of ready-made checklists covering smoke tests, signup and activation, checkout, pricing and landing pages, form validation, accessibility basics, mobile responsiveness, content clarity, navigation and search. Pick one, point it at a URL and adjust the steps, rather than writing every check from scratch.

Map a Slack channel for design review, then drop a link to a page or a Figma file into it and ask for feedback. Stunt Double opens the link, captures what it sees (Figma frames come through your workspace Figma connection) and posts a review back in the thread, with findings ordered by severity. The channel does not need to be wired to a project first.

Resend an invitation that never got opened straight from Settings > Members, and let Stunt Double send a reminder on its own if one is still sitting unaccepted.

Domain auto-join is no longer limited to a single workspace per domain, so teams running separate workspaces on the same company domain can each pick up new teammates automatically.

The public comment widget went through a security pass: every endpoint now checks the requesting origin against the project allow list and fails closed, team-only comments are unlocked by a signed short-lived token rather than a client-side flag, and anything a visitor submits is validated before it reaches your dashboard.

A dedicated product page walks through what actors, checklists and interviews actually do, and the security page now covers how comment data is handled, with the full detail available on request.

Background work was right-sized and rescheduled so scheduled sweeps stop doing no-op work, embeddings are cached per workspace so repeated interview and search queries are not recomputed, and a batch of benign auth and rendering warnings no longer surfaces as errors.

Owners can nominate a business email domain from Settings > Members. Anyone who then signs in with a verified address on that domain joins the workspace automatically, no invitation needed. It stays gated on a verified email and on an owner still being on that domain, so it fails closed if either changes.

Stop an in-flight interview run the moment you have seen enough, instead of waiting for every participant to finish. Whatever has completed stays available.

Point interview participants at a self-hosted worker to run studies against prototypes and sites behind your VPN or SSO, the same way checklists already could.

Integration and validation guides move out of the docs sidebar into a dedicated guides section that reads and shares like articles, with visuals, reading times, and social cards. Old guide links redirect to their new home automatically.

The Slack and Linear agents now run as durable background sessions, so a long-running request keeps going and recovers cleanly instead of dropping partway through.

The macOS app remembers its window position and size, starts without a flash of the wrong theme, supports zoom, moves faster between views, and its native menu now matches the app structure.

The app now recovers on its own from stale-deployment chunk errors after a release, retries transient database hiccups instead of surfacing them, and offers passkey sign-in only where it actually works.

Runner, the app-wide assistant (Cmd/Ctrl+I), now persists your conversation in the workspace, so closing the panel or refreshing the page does not wipe your thread. Start a new chat any time from the panel header.

Inside the macOS app, Settings > Workers now offers a "Use this device" card that drives actors through your own local browser in a dedicated profile instead of a hosted one. Log into VPN or SSO once and every actor task on that worker inherits the session, so Runner and your actors can reach internal apps and sites without a separate CLI or Docker setup.

Choose push alongside email and Slack in notification preferences. Activity that matters can reach you in the browser even when Stunt Double is not the focused tab.

When someone marks your pinned comment resolved, you get an inbox item (and a push or email, depending on your preferences) so closed loops do not get lost.

Visitors on sites with the feedback widget can press C to start placing a pin without reaching for the floating button. Toggle the shortcut off per project from Comment settings if it clashes with your own page shortcuts.

Pins now record their anchor element and the viewport scroll position when placed, so they stay on the spot the poster clicked even when images load late or the layout reflows.

Checklists, interviews, insights, chat, and the workspace assistant now run on Claude Opus 4.8 through Vertex AI, with prompt caching kept on every multi-step agent loop.

We tightened data-handling access controls, restored hot-path database indexes, and trimmed redundant queries and re-renders so the app feels snappier and your workspace data stays better isolated.

The macOS app window is draggable from the title bar again, and automatic updates fetch releases more reliably.

Feedback and comments now have their own top-level view in every project instead of living in a settings tab, with faster triage and resolved comments tucked out of the way until you need them.

Reply directly to a pinned comment to keep the conversation in context, and get an email when someone replies to yours. The composer now protects your draft if a submission fails, so you never lose what you typed.

The visual editor for checklists, automations, and interviews now opens as a true full-screen takeover, giving the canvas the whole window to work with.

Signed in on your phone and your laptop at once? Your own cursor now shows up consistently across every tab and device you have open, alongside the live cursors you already see from teammates.

Start a self-hosted worker with `--browser system` and it drives the machine's own Chrome, headed, in a dedicated profile, instead of a bundled browser. Log into VPN or SSO once in that window and the session persists across tasks and restarts, so actors can reach sites gated behind corporate access without leaving your network. Settings > Workers walks through setup and verifying the connection.

Workers now ship as a single Docker image with no separate install path to maintain, and auto-updates run on a properly maintained Watchtower fork instead of an abandoned one. The Workers page also stopped occasionally showing a worker twice right after you create it.

Managing who is in your workspace is now a dedicated view instead of a tab buried in settings, so invites, roles, and removals are easier to find.

Checklist and interview inspectors now let you edit a start step properties directly, and automations gained the same inline editing for any node, so you rarely need to leave the canvas to tweak a step.

The welcome tour is trimmed down to core concepts and connecting Claude, and onboarding progress is now tracked per person rather than per workspace, so it will not resurface for teammates who have already completed it.

Interview listing cards now surface more at a glance, and we cached embeddings and moved AI Gateway calls through Vertex to make AI features and page loads noticeably snappier across the app and marketing site.

Automations now lay out on a canvas instead of a flat list: see the trigger, every step, and how they connect at a glance. A condition can branch, sending its yes and no outcomes down separate paths, so one automation can handle more than a single straight line. Your existing automations carry over and run exactly as before.

Finished a checklist run worth showing off? Turn on a private share link and anyone can open a read-only report (summary, checks, evidence, and screenshots) without an account, or email the report to up to 20 people in a couple of clicks. Switch the link off any time to revoke access.

Link a Vercel project and point a checklist or interview at its newest preview or production deployment instead of a fixed URL. Each run resolves to the latest build automatically, and access behind Vercel password or login protection is handled for you, so actors never hit a sign-in wall. A "Test access" button confirms it can reach the deployment before you launch.

The visual editors for checklists and interviews are out of preview and available to every workspace. Edit alongside your team with live cursors and presence, and watch run progress stream onto the canvas in real time.

The Workers settings page now shows each worker connect live, with the status dot flipping the moment a heartbeat lands, no refresh needed. Set a network access policy per worker (allow everything, allow only certain domains, or block specific ones) to keep firewalled deployments locked down.

GitHub, Linear, Vercel, Slack, and Figma are all available to connect from one place, with no feature flags or "coming soon" labels in the way.

Cancel a checklist run and it now shows a distinct "Cancelled" state with a confirmation toast, instead of looking like a failure. Choosing where an actor browses from is a searchable, region-grouped picker, so finding the right city is quick even with a long list.

When someone accepts your workspace invitation, you get an email letting you know they are in, so you are not left wondering whether it landed.

We hardened the plumbing behind connected tools: more reliable MCP sign-in, and GitHub, Linear, and Vercel webhooks that keep flowing without dropped events, so your integrations keep working quietly in the background.

Drop in a URL on our home page and watch a full user interview run live, no account needed. The orchestration steps, the live browser, the actor reasoning and responses, and a synthesised report all come together on a single realtime canvas. Sign up to unlock the full report.

Generate a complete interview discussion guide from a short brief, or regenerate one you already have. When a guide exists you choose how the run reconciles with it: enrich it in place, append fresh sections, or replace it entirely.

Pull down from the top of any list or detail view on your phone to reload the latest. The native gesture you already expect now works across the whole app, with the chat thread and canvas editors left free for their own gestures.

Your account page is now split into clear tabs (General, Notifications, Security, and Feedback), so profile, appearance, time zone, passkeys, and two-step verification are easier to find. Each tab is linkable and survives a refresh.

Stunt Double AI features now run through a unified AI gateway with automatic prompt caching, so multi-step work like interviews, checklists, insights, and triage is quicker.

A fresh behind-the-scenes post covers how we build Stunt Double on Supabase. Backstage now sits in the top navigation too, so the blog is one click away from anywhere on the site.

We trimmed first-load JavaScript on the home and blog pages and tuned the in-app experience, so pages feel quicker to open. Floating menus and the workspace switcher now sit on a near-opaque surface, so text never shows through.

Every date and time now shows in your own time zone, detected automatically with an override you can set in Settings from a searchable list with live offsets. Scheduled automations run on the time zone you choose too, so they fire at the right local time.

New tools let you create and manage checklists, automations, and projects directly from connected AI clients like Claude, building on the pull request tools from the last release.

Plans and pricing now display in your local currency based on your location, so what you see is what you pay.

Switching between views no longer triggers a full reload, and each view has its own tailored loading state, so moving around Stunt Double feels noticeably quicker.

Project cards now show each site favicon, project settings are grouped into clear tabs, and the settings link sits at the bottom of the sidebar where you would expect to find it.

Sign in without a password using a passkey (Face ID, Touch ID, or your device PIN), or continue with your GitHub or Figma account alongside Google and email code. Register and remove passkeys any time from your account settings.

Checklists and interviews now have a visual editor. See a checklist laid out from start, through each check, to the report, or map an interview as sections and questions feeding into your participants and findings. Add, edit, and reorder steps on the canvas, then save. Available as a preview.

Choose your workspace language, British, American, or Australian English, from Settings. Generated reports and synthesised findings are now written to match, so the output reads the way your team writes.

Kick off a workflow the moment a pull request is merged, and drive reviews from your own tools with new MCP tools to comment on, list, and read pull requests. Installing the GitHub App and linking repositories now works reliably end to end.

URL fields across project, workspace, and interview setup now bring up the URL keyboard on mobile and add https:// for you, so pasting or typing an address just works.

Bring your stack into Stunt Double from one unified connections view. Link Vercel, Slack, GitHub, and Linear, manage them all in a consistent place, and let webhooks keep everything in sync, so actors can act on the events that matter and surface work back into the tools your team already uses.

Connecting an integration now confirms success or explains what went wrong with a clear message, instead of quietly returning you to the page.

The Index now grades sites against the current agent-readiness landscape with more accurate evidence and fewer false findings, and scores faster too. Every scorecard gains social proof (sector rank and percentile), a working share button, and a direct path to claim your domain and start improving it.

We have published our approach and commitment to security: data protection, access control, compliance, and how self-hosted workers keep firewalled deployments locked down.

Backstage is the new Stunt Double blog: product updates, practical guides, and behind-the-scenes stories from the team, all in one place. Follow along to get more out of your cast and see what we are building next.

A new integrations page covers every way to connect Stunt Double, with a downloadable .mcpb bundle that adds the MCP server to Claude Desktop in a single click. Manual setup for Claude Code and Cursor or Windsurf is a tab away.

Setting up and launching an interview round is quicker. Launching now drops you straight on the Report tab with live progress as sessions complete, and the regenerate control moved up to the page header where it is easy to find.

The sign-in screen is cleaner and quicker to scan, with copy that matches the passwordless code flow rather than the old magic-link wording.

Fixed a bug in GitHub integration authentication so connecting a repository works reliably.

The workspace and project navigation are now organised around three jobs to be done: Discover (interviews and research), Validate (checklists, design reviews, site topology), and Improve (insights, goals, comments). One vocabulary across the sidebar, mobile nav, command bar, breadcrumbs, and the welcome tour.

Every project overview now groups its summary, recent activity, and quick actions under Discover, Validate, and Improve, so the dashboard mirrors the way you actually work, and getting from "what happened?" to the next session takes a single click.

The workspace dashboard surfaces project summaries with the same Discover/Validate/Improve grouping, so you can scan progress across every project without opening each one.

Run Stunt Double browser workers behind your own firewall with a single `docker run`. The image is published as a multi-arch build (linux/amd64 + linux/arm64) on the GitHub Container Registry, so Apple Silicon and Intel hosts both get a native binary.

The recommended setup ships with a Watchtower sidecar that polls every five minutes for new worker images and recreates the container in place. New hosts get auto-update out of the box; existing ones get it the next time you copy the snippet from workspace settings.

Pick a city in your actor settings and every browsing session that actor runs (chat, checklists, insights, topology, interviews) is routed through a Browserbase residential proxy in that location. Useful for testing geo-gated content, localised pricing, and region-specific UX.

Every interview round is now kept on the timeline. Re-run the same discussion guide against a new iteration of your product or Figma file and the previous rounds stay in place, so you can compare findings across versions instead of overwriting them.

Cast members can now carry what they told you in earlier interviews into later sessions and chats. Opt in per interview, and follow-up rounds build on prior context instead of starting from a blank slate.

Design Reviews moved into the Validate section as a tab alongside Checklists and Topology, so all the ways to pressure-test a build live in one place.

Ideas, Goals, and Comments are now one section ("Co-create" inside Improve) with a shared tab bar, so capturing feedback, setting goals, and generating insights feel like one feature rather than three.

The Figma OAuth app is ready for the Figma marketplace and the feature flag is gone, so every workspace can connect Figma and create projects from a file without waiting for an invite.

Prompt caching is now wired through every multi-step Anthropic tool loop (checklists, interviews, insights, chat) with hit-rate instrumentation. Long agent sessions run noticeably faster.

Web, desktop, and the MCP server now ship under a single version number. When you read about v1.5 in the changelog, you know it matches the build you have installed and the server your agent is talking to.

Recruit a panel from your actors or freshly generated personas, build a discussion guide of questions and think-aloud tasks, and launch a round of moderated interviews. Watch transcripts stream in as each session runs.

Describe an audience and Stunt Double generates a persona, with a name, background, and point of view, to round out your interview panel in seconds.

Point an interview at a Figma prototype and participants react to the design frames, so you can validate direction with no live product required.

Every round wraps with a report that surfaces key themes, verbatim quotes, and recommendations ranked by severity. Findings you can act on, not just transcripts to read.

Re-run the same discussion guide on each new version and compare findings round over round to see what actually improved.

New MCP tools let you create an interview, build its discussion guide, launch a round, and read back the synthesised report straight from your AI client or scripts.

Add an authenticator app as an optional second factor on top of email code and Google sign-in, for an extra layer of account security whenever you want it.

Claim a domain on the AI Agent Readiness Index and it now appears as a project in the app, connecting your public score to the work you do to improve it.

Google Gemini is now scored as a live provider on the AI Agent Readiness Index, alongside Claude and ChatGPT, so comparisons reflect the assistants your customers actually use.

Sign-in, invitation, and notification emails now carry the Marquee look with full dark-mode support, so every message looks right whatever your mail client is set to.

Tidy up your account by archiving workspaces you no longer need. In-flight runs wind down cleanly and the workspace disappears from view, with our team able to recover it if you change your mind.

Project settings now live in the sidebar so they are reachable from any page, and switching workspaces on mobile is faster.

A new public leaderboard at index.stuntdouble.io that scores hundreds of real websites on how well they work for AI agents. Compare providers side-by-side, explore scoring methodology, and see how your domain stacks up.

Domain owners can now sign in, edit metadata, re-run scoring on demand, and manage how their site appears on the Index.

A new visual identity rolls out across the app, marketing site, and desktop shell: refined typography with our new Untitled Sans and Serif fonts, brand icons, a warmer palette, and more considered hover and surface states.

Sign-in, invitations, and notification emails now match the Marquee look, so every touchpoint feels like one product.

A fresh cut of the macOS desktop app shipping the latest product work, including the Marquee refresh, improved project dashboard, and pull-to-refresh on touch devices.

On phones, the Index leaderboard now sits above the stats rail so the rankings are the first thing you see.

Run a design review across any page or prototype and watch multiple actors annotate it from their own perspective. Feedback lands on a shared canvas where you can pan, zoom, resolve comments, and re-run the review after making changes.

Create a project directly from a Figma file. Connect Figma via OAuth and bring your designs into Stunt Double so actors can review screens before you ship a single line of code.

Visualise how the pages of your product connect to each other. Stunt Double crawls your site, including links tucked away in the footer, and builds an interactive map you can explore.

Stunt Double now follows your operating system theme by default, and you can override it any time from your account settings. Dark mode, light mode, or whatever your Mac is doing right now.

The mobile drawer now mirrors the full desktop sidebar, so every navigation option is a tap away when you are working from your phone.

Workspace activity (new feedback, completed checks, fresh insights) now streams in more reliably and with less lag, keeping everyone on the same page without needing to refresh.

Page skeletons now match the layout you are about to see, so navigating between sections feels instant and less jarring.

When you open an insight, its parent project stays expanded and the Insights nav item is highlighted, so you always know where you are.

A signed and notarised native macOS app. Runs as a universal binary on both Apple Silicon and Intel Macs, with automatic OTA updates.

A download button in the navigation always points to the latest release. No manual version tracking needed.

Submit feedback from anywhere in your workspace. The form is always accessible, so your team can share thoughts without leaving what they are working on.

See the status of every piece of feedback you have submitted directly from your account page. Know what has been acknowledged, actioned, or resolved.

Reduced from 32 MCP tools to 23 without losing functionality. Cleaner interfaces, fewer redundant calls, and faster agent interactions.

Creating a checklist from a project now prefills the URL and name automatically, saving setup time.

Create goals directly from insight reports and dynamically link them as new findings come in.

Design system reviewer, security auditor, legal reviewer, and compliance checker join the cast, expanding what actors can specialise in.

Resolved an intermittent issue where an error was displayed during sign-in code entry.

Run browser workers behind your own firewall or VPN. Available on Enterprise plans.

Link your GitHub repositories to workspaces and bring actors into pull request reviews. Stunt Double can now surface user perspective directly in your code review workflow.

Connect your Linear workspace and use Stunt Double as an agent inside Linear. When an issue is assigned to an actor, it gets to work automatically.

Connect Figma via OAuth and let actors interact with your designs before a single line of code is written.

A native Stunt Double desktop app for macOS. Now available for download.

Regenerate AI insight reports after making changes, with a full version history so you can always compare what changed and why.

Assign and configure self-hosted workers directly from the workspace dashboard.

Click anywhere on a live prototype or product to drop a pin, write a comment, and start a thread. Replies, notifications, and screenshot attachments are all included.

Reply to any feedback comment and get notified when someone responds. Keeps conversations in context rather than scattering them across tools.

Comments are pinned to specific page versions using a content hash, so feedback stays anchored even as pages evolve.

The chat assistant can now query your projects, goals, issues, checklists, and feedback in real time, giving answers grounded in what is actually happening in your workspace.

View feedback for a specific page or route, including full support for single-page apps that update the URL without a full reload.

Optimised font loading and asset delivery. The app opens noticeably faster, especially on mobile.

The mobile bottom bar now surfaces project views as primary tabs, making it much faster to move between Goals, Checklists, Feedback, and Workflows on small screens.

AI-generated insight reports synthesise feedback and research into structured findings. Now available to all users as a public beta.

Connect individual feedback items and insight reports to product goals, so you can track how user research supports (or challenges) what you are building towards.

Prominent stats, a two-column layout, and an activity feed give you a clear overview of what is happening across a project at a glance.

The goal detail view now uses collapsible sections and inline feedback search, making it easier to focus on what matters when working through a goal.

See a summary of every project directly from the workspace home, without clicking into each one.

The feedback feature is now available to all users.

Sign in and sign up are now the same step. Enter your email or use Google, and an account is created automatically on first use. No separate signup page needed.

A new split-panel layout on login and signup, with feature highlights on the left and the sign-in form on the right. Cleaner and more confident.

The feedback dashboard now has three goal-oriented views: onboarding for first-time setup, page-grouped review for regular use, and dedicated settings.

Gather feedback from each session in one place, attached to the project it came from, so the signal stays close to the work it is about.

A first Model Context Protocol server lets you drive actors, projects, and checklists from your own AI client or scripts.

Get notified when a long-running session finishes or when there is something waiting on you, so you do not have to keep the tab open.

Actors run inside a hosted browser session and act on live pages using natural-language instructions, so they can work through real products rather than just talk about them.

Every step is grounded in a screenshot of the current page, so an actor sees what you would see before it clicks, types, or navigates.

Actors chain actions across a session, reading the result of each step before deciding the next, which makes longer tasks far more reliable.

Describe what should be true about a page or flow and let an actor verify it, so routine checks happen without you driving every click.

Point a checklist at any address and kick off a run, with the URL and name prefilled from the project where it lives.

Each run reports every step with its outcome and supporting detail, so you can see exactly what passed, what failed, and why.

A workspace gives your team one shared home for actors and projects, with members and settings in a single place.

Group related work into projects so each product or initiative has its own space for actors, sessions, and results.

Invite teammates into a workspace by email and start collaborating on the same actors and projects right away.

Spin up an AI actor with its own name, background, and point of view, the foundation everything else in Stunt Double is built on.

Hold a back-and-forth conversation with an actor to pressure-test ideas and hear a perspective that is not your own.

Stunt Double opens to a small group of early users to shape where the product goes next.