AI improvement prompts

دا څنګه کار کوي

په click کې د data دوه pieces combine کېږي:

• Finding — title، description، evidence، remediation، CWE — له report سره مخکې loaded.
• ستاسو د codebase framework — د scan له discovery.tech-fingerprint findings څخه detected کېږي (Next.js، React، Vue، Django، Express، Rails، Laravel، Flask). کله چې framework detected نه شي، generic recipe ته falls back کوي (agent prompt بیا له LLM څخه د repo state له مخې detect کول غواړي).

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 څنګه ښکاري

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

موږ framework-specific snippets د دې لپاره surface کوو:

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

Framework detection best-effort دی. موږ __NEXT_DATA__ tags، __NUXT__، hash cookies (laravel_session)، X-Powered-By headers، او څو نور signals sniff کوو. که تاسو custom framework چلوی، prompt generic recipe ته falls back کوي او agent یې ستاسو له package.json څخه figure out کوي.

له خپل AI agent څخه یې وکاروئ

که تاسو MCP server wire up کړی وي، هماغه prompt د slash command په توګه exposed دی. له Claude Desktop څخه:

/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.

ولې هر click کې Claude نه hit کوو

د launch پر مهال موږ فکر وکړ چې د هر click لپاره Anthropic API call کړو څو prompt له codebase context سره refine شي. مو ونه کړ، ځکه:

• هغه agent چې user یې paste کوي مخکې له مخکې codebase context لري — دوی Cursor / Claude Desktop له خپل repo open سره کاروي.
• د هر (check × framework) templating د value شاوخوا ~80% بې له per-click cost پوښي.
• “Refine with AI for my codebase” opt-in وروسته API fire کولای شي که users یې وغواړي. نن نه.