AI improvement prompts

Wéi et funktionéiert

Zwou Date-Stécker komme beim Klick zesummen:

• De Finding — Titel, Beschreiwung, Evidenz, Remediatioun, CWE — ass scho mam Rapport gelueden.
• D'Framework vun Ärer Codebase — erkannt aus de discovery.tech-fingerprint-Findings vum Scan (Next.js, React, Vue, Django, Express, Rails, Laravel, Flask). Fält op e generescht Rezept zeréck wann kee Framework erkannt gëtt (den Agent-Prompt freet den LLM dann et aus dem Repo-Zoustand z'erkennen).

Templates live in lib/scanner/fix-prompts.ts. The registry has check-specific guidance for crawlability, search presentation, semantic content, structured data, media, performance, accessibility, forms, mobile/i18n, runtime, owner journeys, and repo-quality issues. For everything else, the existing remediation field on the finding becomes the generic recipe.

Wéi de Prompt ausgesäit

Fix the "Hero image is lazy-loaded and missing dimensions" finding on /pricing.

Issue: The largest above-the-fold image is marked loading="lazy" and has
no explicit width/height. That can delay LCP and create layout shift.

Codebase context: Next.js.

Recommended fix:
Use next/image or the existing image component with explicit width/height,
responsive sizes, meaningful alt text, and priority/fetchPriority for the
first major visual on the page.

Constraints:
- Don't break existing tests; run the test suite after the change.
- Match the codebase's existing style and lint config.
- Add a brief comment explaining the performance reasoning only where the
  fix would otherwise look arbitrary.
- If the fix needs a new dependency, install it via the project's
  package manager (npm / pnpm / pip / bundle / composer).

Reference: Core Web Vitals / Largest Contentful Paint guidance.

Ënnerstëtzt Frameworks

Mir weisen framework-spezifesch Snippets fir:

• Next.js, React, Vue, Nuxt, Svelte (Frontend)
• Express, Fastify (Node.js Backend)
• Django, Flask (Python)
• Ruby on Rails
• Laravel (PHP)
• ASP.NET Core (geplangt, haut Fallback op generesch)

Framework-Erkennung ass best-effort. Mir sniffen __NEXT_DATA__-Tags, __NUXT__, Hash-Cookien (laravel_session), X-Powered-By-Headeren an e puer aner Signaler. Wann Dir e personaliséierte Framework leeft, fält de Prompt op dat generescht Rezept zeréck an den Agent fënnt et aus Ärem package.json eraus.

Vun Ärem AI-Agent benotzen

Wann Dir de MCP Server agebaut hutt, gëtt dee selwechte Prompt als Slash-Command exposéiert. Aus Claude Desktop:

/fixweb-fix finding_id=550e8400-e29b-41d4-a716-446655440000

The renderer looks up the finding, detects the framework from the parent scan when available, renders the templated prompt, and injects it into your conversation as the user message. No round-trip to our Claude API; templates are pure and free.

Firwat mir Claude net pro Klick uruffen

Beim Launch hu mir iwwerluet, fir all Klick d'Anthropic API unzeruffen, fir de Prompt mat Codebase-Kontext ze verfeineren. Mir hunn et net gemaach, well:

• Den Agent, an deen de Benotzer paste't, huet scho Codebase-Kontext — si benotzen Cursor / Claude Desktop mat hirem oppene Repo.
• Templating pro (check × framework) deckt ~80% vum Wäert ouni pro-Klick-Käschten.
• En “Refine with AI for my codebase”-Opt-in kéint spéider d'API ausléisen, wann d'Benotzer dat wëllen. Haut net.