agentsclimarketplace

Skill

Skill grupomidiasystems/auditoria-seguranca-lovable-supabase/skill

Use this skill whenever the user wants to audit, review, or harden the security of a project built with Lovable + Supabase (or similar vibe-coded / AI-generated stacks) BEFORE publishing it. Triggers include any mention of "Lovable", "Supabase", "vibe coding", "auditoria de segurança", "security audit", "antes de publicar", "review security", "RLS", "Row Level Security", "secrets exposed", "IDOR", "SSRF", "webhook Stripe", "edge function security", or when the user uploads a Lovable/Supabase project and asks for a security pass. Also use when the user explicitly asks to "rodar o checklist", "verificar segurança", or pastes the AUDITORIA-SEGURANCA-LOVABLE document.From its SKILL.md

Install
npx -y skills add grupomidiasystems/auditoria-seguranca-lovable-supabase --skill skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as CC-BY-4.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

4.4 KB, 871 tokens by cl100k_base, as published. Nobody here has run it

Lovable Security Audit Skill

Comprehensive, opinionated security audit checklist for projects built with Lovable + Supabase, optimized for solo founders, indie hackers, and software houses shipping vibe-coded apps.

Authored by Monaco (GrupoMidiaSystems) and licensed under CC BY 4.0.

When to use this skill

Activate when the user is about to publish a Lovable project, has just published one and wants to harden it, or is reviewing a vibe-coded SaaS for security gaps. Most useful before the first paying customer touches the app.

How to use this skill

The full reference document is ../AUDITORIA-SEGURANCA-LOVABLE-v1.3.0.md (in the repository root). Read that file before performing the audit — it contains the master prompt, 19 modular prompts, manual checklists, SQL diagnostics, curl commands, and prêt-à-coller prompts in Portuguese for the Lovable chat.

Recommended flow

  1. Read the reference .md in full.
  2. Identify the project context: standalone web app, multi-tenant SaaS, mobile/Capacitor, AI-heavy, payment-enabled, etc. The reference document has a "tipos de projeto" matrix mapping which modules to prioritize per project type.
  3. Run the master prompt (the big code block under "PROMPT MESTRE") against the user's project. This generates a structured audit report covering 13 critical areas: RLS, secrets, auth, IDOR, edge functions, storage, input validation, rate limiting, webhooks, CORS/headers, logs, dependencies, Supabase config.
  4. Run extended modules 14–19 as needed: real-attack testing (anonymous / user A / user B), RPC + SECURITY DEFINER, anti-regression of RLS via migrations, AI/prompt injection, incident response, mobile/Capacitor.
  5. Walk the manual checklist "CHECKLIST PRÉ-PUBLICAÇÃO (Manual)" — checks the user must do in the Supabase dashboard, Stripe dashboard, hosting, and Git that cannot be done from code alone.
  6. Recommend an external scan post-deploy (Vibe App Scanner, Rafter, Aikido, OWASP ZAP).

Output format per finding

  • Status: ✅ OK / ⚠️ Atenção / 🔴 Crítico
  • Findings: file/line where applicable
  • Risk: what an attacker can do
  • Fix: ready-to-paste code or SQL
  • Manual verification: what the user must check outside the codebase

End with an executive summary: total issues by severity, top 3 urgent actions, and an explicit "Safe to publish? YES / NO / WITH CAVEATS" verdict.

Language

The reference document is written in Brazilian Portuguese because the primary audience is Brazilian solo founders and software houses. The skill works in any language — translate findings naturally when the user writes in English; preserve original phrasing when they write in Portuguese.

Important notes

  • Disclaimer: this skill reduces common risks but does not replace professional pentest, legal/LGPD review, or compliance validation for regulated systems (health, financial, legal). Pass this caveat to the user.
  • Do not modify code automatically on the first pass: the reference document is explicit that the first audit is a report, not a refactor.
  • Lovable native scanners: 3 of 4 native scanners (RLS analysis, Database security check, Dependency audit) run automatically before publish; the 4th (Code security review) requires manual click on "Update" in the Security view.
  • JWT in Supabase: access tokens are stateless and remain valid until exp even after logout. Do not falsely claim that logout revokes JWTs.
  • SSRF for arbitrary URLs: when the audited app accepts arbitrary URLs from users (e.g., site auditing tools), allowlist alone is insufficient. Apply the layered defense described in Module 8.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 326,861. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.