An in-app feedback widget with automatic AI distillation. Configure a coding role to propose fixes, run your tests, and open draft PRs with your chosen model.
No credit card. 100 reports a month, on us.
npm install testosaurus-sdk
“I click Pay and nothing happens.”
TypeError:
Failed to fetchCheck the failed checkout request.
A starting point to investigate.Your chosen AI model turns feedback and captured evidence into a summary, reproduction steps and a hypothesis to investigate.
The original message and its available evidence, together in one report.
Automatic distillation makes each report easier to act on. Original feedback, captured facts and AI interpretation stay distinct.
Create a GitHub or Linear issue with the context attached. Then use a configured coding role to propose a change from the report.
Preview the handoff“I click Pay and nothing happens.”
Forest / one size
The screen, alongside the words. Screenshot capture depends on consent.
Forest / one size
A simulated example of an opt-in replay. Inputs are masked by default.
TypeError: Failed to fetchCaptured browser context keeps the report grounded in what happened.
The user reports that the Pay button produces no visible response.
Check the failed checkout request.
An inference to investigate, based on the feedback and console error.AI interpretation stays separate from the original feedback and captured facts.
Send the report to GitHub or Linear, or take it into a coding proposal.
The user reports that clicking Pay produces no visible response.
Suggested reproduction stepsGive the agent a role. Choose its model, allowed files, test commands and budget. Turn the report into a proposed change your team can review.
Beta Initially supports Node.js projects.
The report’s feedback, AI summary and captured context ground the proposal.
Your configured role bounds the work. A draft PR is published after its configured tests pass.
Inspect the diff, request a revision on the same PR and decide what gets merged.
Your scope.A proposed response to the failed request, with a test covering the error state.
src/checkout/**npm test -- checkoutExample output, not a live run. Review the diff and test results before merging.
Connect through MCP so it can read the same reports and retrieve available screenshots or replays.
Inspect this checkout report before proposing a fix.
list_reportsresult“I click Pay and nothing happens.”
RPT–024Filter reports by project, status or severity. Read the details that matter next.
get_reportresult“I click Pay and nothing happens.”
ConsoleTypeError: Failed to fetch
Check the failed checkout request.
Capture depends on consent. AI interpretation stays separate from captured facts.
get_assetresultA short-lived download URL gives your agent access to the attachment.
15 minutesRetrieve screenshots or replay bundles when they are available on the report.
Inspect the example screenshotContext ready for investigation.
A browser SDK that fits your app. Add the package, drop in the script, or hand the setup to your coding agent.
Create a project.
Give your feedback a place to land.
Add your app’s domain.
Save the hostname in project settings to enable submissions.
Use your public embed key.
Copy the project’s pk_… key and initialize the SDK in the browser.
npm install testosaurus-sdkimport { init } from 'testosaurus-sdk';
init({
endpoint: 'https://testosaurus.dev',
apiKey: 'pk_YOUR_PROJECT_KEY'
});<script src="https://testosaurus.dev/testosaurus.js?v=27a499acd3a1"></script>
<script>
Testosaurus.init({
endpoint: 'https://testosaurus.dev',
apiKey: 'pk_YOUR_PROJECT_KEY'
});
</script>Integrate Testosaurus into this app.
Read https://testosaurus.dev/docs/api.html first.
Use the testosaurus-sdk package and its named init export,
or the versioned script supplied by the project dashboard.
Initialize it in the browser with endpoint https://testosaurus.dev
and my project's public embed key (pk_...).
Ask me for the project key if it is not available.
Confirm this app's hostname is saved in the project's Domains.
Use the default consent flow for screenshot and replay capture.
Verify that the widget opens, then let me submit a test report.Choose your AI model, where it can make changes, and how much it can spend.
Screenshots and replay are opt-in. Replay inputs are masked by default.
Capture & consentSet per-run and monthly budgets for each coding role. Runs use prepaid credit; unused reserved credit returns when the run settles.
Understand billingJSON export on every tier, plus an API to build your own workflows.
Explore the APIStart free. Add capacity when you need it.
Keep your feedback, whichever plan you choose.
For your next launch
Start listening. Get the context.
Start collecting freeFor products in motion
More feedback. Room for your team.
Get started with ProNeed more capacity or specific support? Let’s talk about a custom plan
Included AI credit covers report analysis. Coding proposals use prepaid AI credit separately. AI is billed at catalog token rates plus 5%, with inference through Vercel AI Gateway. Full pricing details
Yes. The browser widget works with React, Vue, Svelte, Next.js, or plain HTML. Install the npm package in a bundler app, or add the script to your page. Register your domain and use your project’s publishable key.
Read the quickstartThe user’s message travels with browser context such as the page URL, viewport, language, and console entries. Screenshots and session replays are optional and require consent. The example above shows a report with these attachments enabled.
See capture and consent detailsEnable automatic distillation and choose your model. New feedback becomes a structured report with a summary, category, severity, reproduction steps, and a suggested cause to investigate. The original feedback stays available, and you can collect reports without AI. Free includes €1/month for AI report analysis; Pro includes €3. Your model and report size determine how far that credit goes. On Free, AI analysis pauses when the allowance is spent.
Explore reports and distillationYes. Configure a coding role with your chosen model, allowed files, test commands, and budgets. It can open a draft PR after all configured tests pass. Review the diff, request a revision on the same PR, and decide when to merge. Coding proposals are in beta for Node.js projects and use prepaid AI credit, separate from included report analysis credit.
See the draft-PR workflowYes. Use “Connect an agent” in workspace settings and give its one-use setup prompt to your agent. Through MCP, an authorized agent can configure the project, inspect reports and evidence, and prepare coding roles. You complete GitHub and Linear connections through their normal sign-in and authorization screens. Writable access is required to change settings or start coding runs.
Read the agent setup guideYes. AI uses published model catalog token rates plus 5%, through Vercel AI Gateway. Coding runs reserve their maximum budget from prepaid AI credit and return the unused portion when settled. Set run and monthly coding budgets, usage limits, and alerts. On Pro, opt into additional AI report analysis using prepaid credit. Keep it funded with AI credit top-ups, optional automatic top-ups, or recurring credit on a monthly Pro subscription. JSON export is available on every plan.
Explore usage and billingCapture the evidence. Let AI distill it.
Review the proposed change.
