delegation-quality

To ensure sub-agent delegation returns verifiable output, define atomic tasks with explicit deliverables, constraints, and required evidence. Enforce acceptance criteria so sub-agents return a structured output containing Summary, Files, Verification, and Risks for traceable results.

The best way to structure research delegation in multi-agent workflows is to enforce acceptance criteria requiring sub-agents to return Evidence, Sources, Risks, and Next Steps. This structured delegation approach ensures traceable outputs and clear follow-ups.

You enforce acceptance criteria for delegated debugging and implementation tasks by requiring sub-agents to submit a structured output that includes a Summary, affected Files, Verification steps, and identified Risks before reviewing the deliverables.

Task delegation to sub-agents produces vague deliverables and missing evidence when it lacks disciplined review checkpoints. Applying structured delegation with explicit deliverables and evidence review resolves unclear follow-ups and improves multi-agent reliability.

Yes, you can use structured delegation for verification and review workflows involving multiple agents. The skill governs multi-agent orchestration by enforcing risk assessment, evidence review, and clear follow-ups for each delegated implementation or verification task.

Before delegating atomic tasks to sub-agents, you need to define explicit deliverables, specify the tools and constraints allowed, and establish the evidence required for the task. This setup ensures the delegation output is verifiable and traceable.