seatlisted

# Delegating to Codex The **user** requests the work; the **coordinator** chooses and synthesises the composition; one Codex **seat** performs one deliverable under rights declared in its prompt. Everything below is addressed to the coordinator. What reaches the user is prose you write in their own language: name the agent, say what it did, and say in ordinary words what it may write and where. ## Composition Apply all five rules: 1. Announce the composition **before** starting any Codex run, naming the count and which seats are Codex. 2. Treat refusal as composition: for “no codex” or “just you”, run zero Codex seats and say the resulting panel is all-Claude and shares one model bias. 3. Attribute every finding; if a Codex seat failed or returned nothing, say so and never backfill it with a Claude answer. 4. Knowing the answer is not a reason to skip a requested second opinion. 5. Never add allow-rules on the user's behalf. | What the user says | Composition | | --- | --- | | “no codex”, “just you” | zero Codex seats | | nothing | panels, refutation, competing designs: one dissenting Codex seat; mechanical fan-out or one ordinary task: zero | | “a codex seat”, “one of them codex” | exactly one | | “half codex” | half the seats, rounded up | | “mostly codex” | every seat except the coordinator | | “only codex”, “all codex” | every seat, including a one-agent task | | “two of five codex” | exactly as stated | A dissenting seat pays for decorrelation; mechanical fa