.agents/skills/code-changes/references/delegate.md
Match the executor to the task, not the other way around. The point is cost and speed without losing quality — the coordinator stays accountable for the result.
Tier definitions and per-vendor model examples: model-tiers.md.
Task profile: Trivial: mechanical edit, config tweak, typo, doc update
Task profile: Standard, well-specified implementation
Task profile: Hits an escalation trigger (see escalate.md)
executor_tier value (see
artifacts.md); a task's executor_tier stays trivial, implementer, or
coordinator-direct even when an escalation was used to unblock it.The coordinator typically runs at implementer tier itself, so standard tasks are usually executed directly rather than delegated. Delegate anyway when there's real parallelism — independent tasks, each in its own worktree — or to shed trivial mechanical work onto a cheaper model.
Escalation is never the default path for "ambiguous" or "architectural" — it fires only on the concrete triggers in escalate.md, and only for the specific question, not the whole task.
Hand the subagent its full pinned spec from Phase 2, plus:
Analysis, final verification, and delivery never go to an implementer — they stay with the coordinator or go up to Escalation tier for a specific question: