5 tagged devils advocate, measured the same way as everything else here.
Skill Claude CodeCodex
ES — Critica a fondo una idea, plan, pitch o proyecto entero sin validar ni adular: asume que va a fracasar y lo demuestra por ocho ángulos (premisas, mercado, competencia, viabilidad, números, ejecución, pre-mortem y punto ciego), investiga fracasos reales cuando hay herramientas, y entrega un veredicto franco con la…
Skill Claude CodeCodex
A structured critique of proposals to create an AI imitation of a colleague from their chat messages. It explains why chat history may not capture a person’s knowledge, decisions, or changing behavior.
Skill Claude CodeCodex
Use when the user wants a debate, second opinion, adversarial review, or iterative discussion between Claude and Codex (the OpenAI CLI) about a spec, design, plan, document, position, refactoring, or codebase — the subject can be anything, technical or non-technical, e.g. "ask Codex", "debate this with Codex", "have…
Skill Claude CodeCodex
Use ONLY when the user explicitly asks to build a project-specific war-room roster (e.g. via /war-room-roster). Scans the project to infer its domains and proposes a .war-room/roster.yaml of domain officers. This is opt-in and heavier than a normal war-room run — never run it automatically, and never write the roster…
Skill Claude CodeCodex
Use when a decision has no right answer yet — a design fork, an architecture call, a hard tradeoff with no test or spec to check against. Convenes a council of non-colluding officers plus a mandated dissenter, and produces a winning strategy with the failure-worlds that would beat it. Do NOT use when an oracle (a…