dispatching-parallel-agents

Group failures by independent root cause, then dispatch one subagent per problem domain in a single response so they run concurrently. Each agent gets a focused prompt with the failing test names, error messages, constraints, and an expected summary of findings.

Use parallel dispatch when failures are independent, such as different test files or subsystems with no shared state. Avoid it when failures are related, when understanding requires full system context, or when agents would edit the same files.

A good prompt is focused on one problem domain, self-contained with pasted error messages and test names, explicit about constraints like not changing production code, and specific about the expected output such as a summary of root cause and fixes.

Yes, if agents share state or edit the same files their changes can conflict. The pattern requires independent domains, and after agents return you should review summaries, check for overlapping edits, and run the full test suite to verify integration.

Multiple dispatch calls issued in the same response execute concurrently, while dispatches sent one per response run sequentially. Batching them in a single response is what enables true parallel execution and time savings.