Ask for the screenshot of a named state and a named run destination. An unspecified screenshot can show a launch screen, stale build or unrelated simulator.
- 1Named state
- 2Reproduction steps
- 3Capture
- 4Compare
Work through the example
Prompt: Capture the empty-search state after entering a query with no matches. State the runtime and build you launched.
One practical example
Use the prompt above on one small, named part of your app. Before accepting an edit, ask the agent to point to the source or configuration that supports its answer. If it cannot locate that context, resolve the missing input before expanding the task.
In the diagram, each arrow represents a handoff you can inspect. Keep the intermediate result rather than jumping directly to the final claim. For example, a reported finding should retain its location, and an executed check should retain its outcome. This gives the next attempt a concrete starting point and makes a misleading completion message easier to challenge.
Apply it to your project
Prompt: Capture the empty-search state after entering a query with no matches. State the runtime and build you launched. Then compare the resulting diff with the original request. If it changed a neighboring concern, ask why that change was necessary. Save the result only after the relevant check has run, and record any part that remains unverified.
Evidence and limits
A screenshot is visual evidence, not proof of every behavior.
This is an educational guide. Its presence in the series does not certify a completed client-specific lab. The series evidence record separates executed checks from exercises and blocked environments.
Inspect the source used in this lesson.
Related reading
- Five checks for an AI-built SwiftUI app
Use persistence, search and accessible states to make “it works” a testable claim.
What to do next
Next: Pin your Xcode version in AGENTS.md and stop the agent guessing APIs