delegate

name: delegate description: Hand a task to another AI agent peer over the repowire mesh — reuse an existing peer on the chosen backend, or spawn one if none exists. Use when you want to offload work to a specific agent (e.g. delegate a refactor to a Codex peer) and track it to completion.

Delegate a task to a peer

Hand a self-contained task to a peer on a chosen backend and track it via the ask/ack lifecycle.

Resolve the delegate backend

1. Explicit argument — if the user named a backend, use it.
2. Configured default — else:

   repowire config get skills.default_delegate_backend
   

3. Safe fallback — else ask the user which backend to delegate to. Don't hardcode one. (Unlike review/plan, delegation MAY target your own backend — delegating to another claude-code peer is legitimate.)

Choose / create the executor

1. Prefer an existing peer on the chosen backend (idle if possible):
   • MCP: list_peers() → pick an online peer with the right backend.
   • CLI: repowire peer list.
2. Spawn only if none exists AND the user is clearly delegating (don't spawn silently). Confirm before spawning:
   • MCP: spawn_peer(path, backend=<chosen>, circle=<skills.default_circle or current>)
   • CLI: repowire peer new ... Default circle: repowire config get skills.default_circle (else your current circle).

Hand off + track

1. Send a complete, self-contained brief (the peer doesn't share your context):
   • MCP: ask(peer_name, "<full task brief + acceptance criteria>")
   • Tracked delegation is MCP-only (repowire peer ask is a synchronous test utility, not the ask/ack lifecycle). To close from the CLI: repowire peer ack <cid>.
2. ask returns a correlation_id; the peer reports back via ack(corr_id, ...).
3. Track to completion; chain follow-ups with ask(reply_to=corr_id, ...).

Use notify_peer (fire-and-forget) only for nudges, not for tracked delegation — delegation needs the ack lifecycle.