silvioventre
- 8 skills
- 0 followers
- 6 hours ago last updated
- ▌ Skill Name In Kebab Case · silvioventreUse when <concrete trigger — when Lovable should load this skill>. Describe the boundaries too, meaning what this skill does NOT cover.
- ▌ Ship · silvioventre bundleUse when the user is about to publish, deploy, ship, launch, or go live with a project, or has already published and something about the live site is wrong — "publish this", "make it live", "deploy it", "am I ready to launch", "my changes aren't showing on the live site", "who can see this", "make it private", "share it with my team", "unpublish it". Covers the pre-flight checks before going live, the difference between project access and website access, choosing who can reach the published app, verifying the live site afterwards, and republishing after changes. Not for custom domain DNS setup or workspace role administration, which are configuration the user does in the Lovable interface.
- ▌ Secure · silvioventre bundleUse when the user asks about the security of their app, wants a security review or audit, is about to publish or go live, or is building anything that handles credentials, payments, personal data, roles, or permissions — "is this secure", "review my security", "check for vulnerabilities", "am I safe to publish", "where do I put my API key", "set up RLS", "add authentication", "can users see each other's data", "protect this route". Also use when a change touches secrets, API keys, edge functions, row-level security, login, sessions, or role checks. Routes the concern to the right layer — frontend, edge function, database policy, or authentication — and applies the rule that belongs there. Not for platform account security or general web security theory; this is about the code in this app.
- ▌ Debug · silvioventre bundleUse when something in the project is broken, failing, or behaving unexpectedly — a build error, a red error message, a blank or white preview, a feature that stopped working, a component that disappeared, data that will not load, an edge function or permission failure, an error that keeps coming back after repeated fix attempts, or an app that runs but does the wrong thing. Also use when the user says "it's broken", "not working", "fix this error", "the screen is blank", "it worked before", "I'm stuck in a loop", or pastes a console error, stack trace, or log. Routes the symptom to the matching diagnostic playbook, finds the root cause, and fixes that rather than the symptom. Not for adding new features, and not for performance tuning unless something is actually failing.
- ▌ Responsive · silvioventre bundleUse when the user wants an app, site, page, or component to work correctly on every screen size — "make it responsive", "fix it on mobile", "it breaks on my phone", "it doesn't look right on tablet", "there's horizontal scrolling", "the layout overflows", "make it work on small screens", "test it at different breakpoints", "adapt it to any device". Covers auditing every width from 320px up, diagnosing overflow and reflow failures, applying fluid type and container queries, meeting touch-target and zoom accessibility requirements, and verifying the result against a pass gate. This is execution and verification rather than visual redesign — use it to make a layout actually work at every size, not to decide what the experience should be.
- ▌ Deploy External · silvioventre bundleUse when the user wants to run part or all of their app outside Lovable Cloud, or is weighing whether to — "can I self-host this", "deploy to Netlify or Vercel or Cloudflare", "host it on AWS", "generate a Dockerfile", "move my backend to Supabase", "export my data", "am I locked in", "what happens if I stop using Lovable", "we need this on our own infrastructure for compliance". Covers whether to move at all, what you become responsible for once you do, what migrates automatically versus manually, the build requirements for hosting the frontend elsewhere, and the in-project changes a migration needs. Not for publishing on Lovable Cloud, which the ship skill covers.
- ▌ Art Direction · silvioventre bundleUse when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers landing pages, marketing sites, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, responsive behavior, theming, anti-patterns, typography, spacing, layout, color, motion, micro-interactions, UX copy, error states, edge cases, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, and ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.
- ▌ Lovable Codebase Audit Cleanup · silvioventre bundlePerform a two-phase codebase rationalization on a Lovable project — a comprehensive read-only audit to find dead code, unused code, obsolete dependencies, duplicated implementations, and maintainability risks, followed by a strictly-scoped, approval-gated cleanup of only the batches the user explicitly approves. Use this skill whenever the user asks to "audit the codebase," "find dead/unused code," "clean up the project," "reduce technical debt," "find unused dependencies," references an approved audit batch (e.g. "implement Batch 2"), or asks for a codebase health check, even if they don't use these exact words. Do not skip straight to editing code when this skill applies — always determine which phase (Audit or Cleanup) the request belongs to before acting.