nujovich

Use ANTES de elegir un modelo de inferencia para un job con valor económico. Declarás el revenue esperado y el burn proyectado; Argus decide el tier de compute (ULTRA / BASE / REJECT) y te devuelve el modelo permitido + un auth token que DEBÉS reenviar en tus llamadas a Nemotron. No elijas modelo de memoria: pedí el tier primero y usá exactamente el que Argus asigne.

REQUIRED before any external money spend (Stripe charges, paid API calls, SaaS provisioning, NIM credits, anything that costs real dollars). Argus's financial control plane gates the spend through policy + human approval. Use this for every spend, no matter how small — the policy decides whether it auto-approves or escalates.

Use when the user wants to generate a visual summary / infographic / release card of what ships in a mint release. Reads CHANGELOG.md, editorializes the changes, and renders a mint-branded PNG. Trigger on "release infographic", "release card", "infografía del release", or before publishing a version.

Use when debugging any bug, test failure, crash, or unexpected behavior WITH an AI coding agent — before letting it propose or apply a fix. Stack-agnostic. This skill is about steering the agent through a disciplined loop: forcing it to reproduce and locate before fixing, to show evidence instead of conclusions, and to change one thing at a time. Trigger it whenever the agent starts guessing, proposes a fix without a repro, or claims something is "fixed" without proof.

Notify the user via beep, Windows popup, WhatsApp, and email when the agent needs attention.