A background build needs clear ownership, logs and a completion signal. Do not equate generic process execution with a specific client's subagent feature.
- 1Start build
- 2Retain logs
- 3Wait for completion
- 4Inspect exit status
Work through the example
Prompt: Report the running command and final exit status; do not mark the task complete while the build is still running.
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: Report the running command and final exit status; do not mark the task complete while the build is still running. 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
Antigravity background-subagent behavior is unverified here, so no client-specific command is provided.
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
- How can Claude Code hooks protect generated files and verify a project before stopping?
A tested generated-file guard and Stop check, with configuration, reproducible exit codes, and enforcement limits.
- Muse Code: what our integration actually verifies
Discovery and hooks are useful milestones, but they are not a finished app.
What to do next
Next: Claude Code plan mode before any multi-file Swift change