agent-codex-gatelisted

# Agent + Codex Gate Pattern Spawn subagents to do work, then each agent submits its work to Codex for review via the LLM gateway. Agents iterate on Codex feedback until they get unconditional approval. Work is not accepted until Codex approves. ## Dispatch Defaults Apply these on every dispatch unless the caller has explicitly overridden a rule in the current turn: 1. **Omit `model`** — let the gateway use its configured default per CLI. Nominating a model risks deprecated IDs (`o3`, `o3-pro`, `gpt-4o`, …) and capability mismatches. 2. **`approvalStrategy:"mcp_managed"`** is the skill dispatch default (the gateway schema default is `"legacy"`). For Codex, also pass `fullAuto:true`; this gives sandboxed autonomy while keeping the gateway approval gate in front of execution. 3. **No wallclock timeout; poll every 60 s** — `idleTimeoutMs` is a separate no-output safeguard. 4. **Iterate until unconditional APPROVED** (review dispatches only) — every review prompt must end with "End with APPROVED or NOT APPROVED with findings." Loop: dispatch → poll → parse verdict → on `NOT APPROVED` or conditional approval, dispatch fixes + re-review → repeat. Escalate after 3 rounds. This rule does **not** apply to pure implementation or non-review analysis dispatches. ## When to Use - Dispatching multiple parallel agents to implement different tasks - Any workflow where "spawn agents, let them work, gate on Codex approval" is the pattern - When you want autonomous agents with quality gat