AI improvement prompts

Necə işləyir

Click zamanı iki məlumat parçası birləşir:

• Finding — title, description, evidence, remediation, CWE — report ilə artıq yüklənib.
• Codebase-inizin framework-ü — scan-in discovery.tech-fingerprint findings-indən aşkar olunur (Next.js, React, Vue, Django, Express, Rails, Laravel, Flask). Framework tapılmayanda generic recipe-ə fallback edir (agent prompt sonra LLM-dən bunu repo state-dən aşkarlamağı istəyir).

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.

Prompt necə görünür

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.

Dəstəklənən frameworks

Framework-specific snippets bunlar üçün göstərilir:

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

Framework detection best-effort-dir. __NEXT_DATA__ tags, __NUXT__, hash cookies (laravel_session), X-Powered-By headers və bir neçə başqa signals sniff edirik. Custom framework işlədirsinizsə, prompt generic recipe-ə fallback edir və agent bunu package.json-dan anlayır.

AI agentinizdən istifadə edin

MCP server qoşmusunuzsa, eyni prompt slash command kimi exposed olur. Claude Desktop-dan:

/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.

Niyə hər click-də Claude çağırmırıq

Launch zamanı hər click üçün Anthropic API çağırıb prompt-u codebase context ilə refine etməyi düşündük. Etmədik, çünki:

• User-in paste etdiyi agent-də artıq codebase context var — onlar Cursor / Claude Desktop-u repo açıq halda işlədirlər.
• Per-(check × framework) templating per-click cost olmadan dəyərin ~80%-ni örtür.
• “Refine with AI for my codebase” opt-in gələcəkdə users istəsə API-ni fire edə bilər. Bu gün yox.