Menu
Flag potentially expensive technical choices BEFORE they are committed. Watch for AI model selection, hosting tier upgrades, third-party services with metered pricing, and DB patterns that explode with scale.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Plan post-first-100-users growth without burning out the founder, the budget, or the user base. Pairs with `first-100-users` (which gets you to 100); this skill takes you from 100 to ~1000 sustainably. Activated when the user says "já tenho 100 usuários, e agora?", "como cresço sem queimar dinheiro?", "o WhatsApp 1-a-1 não escala mais".
Get a second (and third) AI opinion on hard-to-reverse decisions. Runs the same prompt through different agent roles or model families in parallel, then reconciles disagreements. Inspired by gstack's `/codex` cross-review pattern. Activated for architecture decisions, security-sensitive code, pricing/packaging changes, deploy gates, legal-adjacent text, or any decision the user flags as "não dá pra desfazer fácil".
Turn a manually-validated workflow into a documented, repeatable, and partially-automatable process. Use AFTER a task has been done at least 3 times by hand and is showing signs of repetition. Inspired by Sahil Lavingia's principle "do it manually until it works, then processize." Activated when the user says "tô fazendo isso toda semana", "vamos automatizar?", "isso já cansou", or after `daily-standup` flags repeated manual work.
Recommend external repo packs from docs/registry/ that fit a project's stack, domain, and constraints. Use during /project-start (after the Technical Plan), when the user asks "que repos eu deveria usar?", or whenever a new need (security review, design system, deploy guidance) appears mid-build.
At the start of a session (or on user request "o que a gente fez", "onde paramos"), produce a 4-bullet briefing — what's done, what's in progress, what's next, what's blocked. Replaces the dev habit of standup for solo vibe coders.
Walk a vibe-coder through deploying a Next.js + Supabase project to Vercel + Supabase production for the first time. Activated when the user says "vamos publicar", "como coloca no ar", "deploy", or after Sprint 1 ships and the app needs a public URL.
| name | cost-watchdog |
| description | Flag potentially expensive technical choices BEFORE they are committed. Watch for AI model selection, hosting tier upgrades, third-party services with metered pricing, and DB patterns that explode with scale. |
Cloud services are like a cab meter. Some run flat-rate (you pay even idle), some run per-trip (per request), some are free up to a limit then suddenly expensive. Before you ship, I'll show you which meter is running and what the worst-case bill looks like if your idea goes viral.
💰 cost-watchdog flags
[HIGH] Using Claude Opus inside a loop that may run 50+ times per user action.
Estimated cost: $0.45 per user action. At 1000 users/day = $13,500/month.
Suggest: switch to Sonnet 4.6 ($0.06/action) or Haiku 4.5 ($0.01/action).
[MED] Realtime Supabase subscription on `messages` table without filter.
Will fire on every insert across all users.
Suggest: filter by chat_id.
[OK] Stripe checkout: percentage-based, no fixed cost.
[OK] Vercel deploy: still on Hobby free tier.