AI improvement prompts

Founga 'oku ngaue ai

Data pieces 'e ua 'oku combine 'i click:

• Ko e finding — title, description, evidence, remediation, CWE — already loaded mo e report.
• Framework 'o ho'o codebase — detected mei he scan's discovery.tech-fingerprint findings (Next.js, React, Vue, Django, Express, Rails, Laravel, Flask). Falls back ki generic recipe pe a 'ikai detected ha framework (pea kole e agent prompt ki he LLM ke detect mei he 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.

Fōtunga 'o e prompt

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.

Frameworks 'oku poupou'i

'Oku mau surface framework-specific snippets ma'a:

• 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 to generic today)

Framework detection 'oku best-effort. 'Oku mau sniff __NEXT_DATA__ tags, __NUXT__, hash cookies (laravel_session), X-Powered-By headers, mo ha signals kehe. Kapau 'oku ke running ha custom framework, 'oku fallback e prompt ki generic recipe pea figures it out 'e he agent mei ho'o package.json.

Ngaue'aki mei ho'o AI agent

Kapau kuo ke wired up e MCP server, 'oku exposed e same prompt ko ha slash command. Mei 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.

Ko e ha 'oku 'ikai ai ke mau call Claude 'i click kotoa

'I launch na'a mau consider calling e Anthropic API ma'a click takitaha ke refine e prompt mo codebase context. Na'e 'ikai, koe'uhi:

• Ko e agent 'oku paste ki ai 'e he user 'oku already ma'u codebase context — 'oku nau ngaue'aki Cursor / Claude Desktop mo honau repo open.
• Templating per-(check × framework) 'oku cover ~80% 'o e value 'ikai ha per-click cost.
• Ha “Refine with AI for my codebase” opt-in 'e lava fire e API later kapau 'oku fiema'u 'e users. 'I he 'aho ni, 'ikai.