A little guidance. A lot less guesswork.

01 / The field guide

Your first check, step by step.

Create a free account and add your app using its deployed website address. Use a site you own and can verify. Private or local addresses are not eligible targets for a browser check.

• Add the ownership proof shown in the app, then verify it.
• Choose Quick Check and review the allowance before starting.
• When the run completes, open the findings and inspect the evidence.
• Choose one actionable issue. Review the suggested fix with your developer or coding tool.
• Deploy the change, then recheck. Compare what was observed.

02 / The field guide

Make each fix specific.

A good fix starts with a clear problem. Read the finding, inspect the evidence and check that it matches your intended behavior before copying the suggested prompt.

Give your developer or AI coding tool the relevant context: the page, what you expected, what happened and any constraints. Review the resulting change. A suggested prompt is not a guaranteed solution.

• Fix one issue at a time when practical, so changes are easier to review.
• Never paste live credentials or customer data into a prompt.
• Test the changed journey and nearby behavior in your own environment.
• Deploy to the correct environment before asking Gauntly to recheck.

03 / The field guide

Before you call it ready.

Use this as a practical starting point, then add the checks your own app needs. Automated evidence only describes the work that actually ran.

• Confirm the deployed version and environment are the ones you intend to release.
• Walk through the main user journeys yourself, including sign-up and recovery.
• Review open findings and decide who owns each remaining issue.
• Check important device layouts; browser emulation does not replace testing on real devices.
• Verify payments, permissions, email and integrations in their appropriate test environments.
• Check that privacy information and support contact details are accurate.
• Review the receipt’s scope and notes before you share it.

04 / The field guide

Prepare a separate place to test.

For Cheater and Hacker, use a staging deployment you control. Verify its exact origin independently of your main app. Set up disposable accounts and the specific scenario you want to test.

Provide the required authentication proof, setup steps and response assertions. Data-changing scenarios need cleanup, and payment scenarios need test mode. Missing setup is not evidence that an app passed.

05 / The field guide

Understand your check allowance.

Think of credits as two balances for running work: one for AI help and one for testing. Different checks cost different amounts. Your account shows the balance and the cost before a check begins.

Plans also cap individual check types. The maximum counts are not all additive: you still need enough credits in the right balance. Paid access currently lasts 30 days per purchase; credit allowances reset each calendar month.