AI improvement prompts

How it works

Two pieces of dati combine on click:

• The risultatu — title, description, evidence, remediation, CWE — already loaded cù u raportu.
• Your codebase's framework — detected da u scansione's discovery.tech-fingerprint risultati (Next.js, React, Vue, Django, Express, Rails, Laravel, Flask). Falls back à a generic recipe quandu nò framework hè detected (u agent prompt then asks u LLM à detect it da u repo state).

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.

What u prompt looks like

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.

Supported frameworks

We surface framework-specific snippets per:

• Next.js, React, Vue, Nuxt, Svelte (frontend)
• Express, Fastify (Node.js backend)
• Django, Flask (Python)
• Ruby on Rails
• Laravel (PHP)
• ASP.NET Core (planned, fallback à generic today)

Framework detection hè best-effort. We sniff __NEXT_DATA__ tags, __NUXT__, hash cookies (laravel_session), X-Powered-By headers, è a few other signals. If voi're running a custom framework, u prompt falls back à u generic recipe è u agent figures it out da u vostru package.json.

Use it da u vostru AI agent

If voi've wired up u Servitore MCP, u same prompt hè exposed as a slash command. From 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.

Why noi don't hit Claude per click

On launch noi considered calling u Anthropic API per each click à refine u prompt cù codebase context. We didn't, because:

• The agent u user pastes into already has codebase context — they're using Cursor / Claude Desktop cù their repo open.
• Templating per-(check × framework) covers ~80% of u value senza any per-click cost.
• A “Refine cù AI per my codebase” opt-in could fire u API later if utenti want it. Today, nò.