AI improvement prompts

Yadda yake aiki

Abubuwan data biyu suna haɗuwa idan aka danna:

• Finding ɗin — title, description, evidence, remediation, CWE — an riga an loda su tare da rahoton.
• Framework na codebase ɗinka — ana gano shi daga findings na scan ɗin discovery.tech-fingerprint (Next.js, React, Vue, Django, Express, Rails, Laravel, Flask). Yana komawa generic recipe idan ba a gano framework ba (sai agent prompt ya nemi LLM ya gano daga yanayin repo).

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.

Yadda prompt yake kama

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 da ake tallafawa

Muna nuna framework-specific snippets don:

• Next.js, React, Vue, Nuxt, Svelte (bangaren frontend)
• Express, Fastify (backend na Node.js)
• Django, Flask (Python)
• Ruby on Rails
• Laravel (PHP)
• ASP.NET Core (an shirya, yana komawa generic a yau)

Gano framework best-effort ne. Muna sniff tags __NEXT_DATA__, __NUXT__, hash cookies (laravel_session), headers X-Powered-By, da wasu sigina. Idan kana amfani da custom framework, prompt yana komawa generic recipe kuma agent yana gano shi daga package.json ɗinka.

Yi amfani da shi daga AI-agent ɗinka

Idan ka haɗa MCP server, ana fallasa wannan prompt ɗin a matsayin slash command. Daga 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.

Dalilin da ya sa ba ma kiran Claude kowane danna

A lokacin launch mun yi tunanin kiran Anthropic API a kowane danna don tace prompt da codebase context. Ba mu yi ba, saboda:

• Agent ɗin da mai amfani yake liƙawa cikinsa tuni yana da codebase context; suna amfani da Cursor / Claude Desktop tare da repo ɗinsu a buɗe.
• Templating per-(check × framework) yana rufe kusan 80% na ƙimar ba tare da kuɗin kowane danna ba.
• Za a iya samun opt-in “Refine with AI for my codebase” da zai kira API daga baya idan masu amfani suna so. A yau, a’a.