// docs / ai fix prompts
AI improvement prompts
Every finding has a Copy prompt button beneath the remediation. Click it, paste into Claude / Cursor / Copilot, and the agent gets the canonical improvement recipe for that website-quality issue — no Claude API call from us.
ଏହା କିପରି କାମ କରେ
Click ଉପରେ ଦୁଇଟି data ମିଶେ:
- Finding — title, description, evidence, remediation, CWE — report ସହିତ ପୂର୍ବରୁ loaded।
- ଆପଣଙ୍କ codebase ର framework — scan ର
discovery.tech-fingerprintfinding ରୁ detected (Next.js, React, Vue, Django, Express, Rails, Laravel, Flask)। କୌଣସି framework detect ନହେଲେ generic recipe କୁ fallback କରେ (ତାପରେ agent prompt LLM କୁ repo state ରୁ ଏହା detect କରିବାକୁ କହେ)।
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 କେମିତି ଦେଖାଯାଏ
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 framework
ଆମେ ଏହା ପାଇଁ framework-specific snippet ଦେଖାଉ:
- Next.js, React, Vue, Nuxt, Svelte (frontend UI)
- Express, Fastify (Node.js backend server)
- Django, Flask (Python framework)
- Ruby on Rails framework
- Laravel (PHP framework)
- ASP.NET Core (planned, ଆଜି generic କୁ fallback)
Framework detection best-effort। ଆମେ __NEXT_DATA__ tag, __NUXT__, hash cookie (laravel_session), X-Powered-By header, ଏବଂ କିଛି ଅନ୍ୟ signal sniff କରୁ। ଯଦି ଆପଣ custom framework ଚଲାଉଛନ୍ତି, prompt generic recipe କୁ fallback କରେ ଏବଂ agent ଆପଣଙ୍କ package.json ରୁ ଏହା ବୁଝେ।
ଆପଣଙ୍କ AI agent ରୁ ଏହା ବ୍ୟବହାର କରନ୍ତୁ
ଆପଣ MCP server wire କରିଥିଲେ, ସେହି prompt slash command ଭାବେ expose ହୁଏ। Claude Desktop ରୁ:
/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.
ପ୍ରତ୍ୟେକ click ରେ ଆମେ Claude କୁ hit କାହିଁକି କରୁନାହିଁ
Launch ସମୟରେ ପ୍ରତ୍ୟେକ click ପାଇଁ codebase context ସହିତ prompt refine କରିବାକୁ Anthropic API call ବିଚାର କରିଥିଲୁ। ଆମେ କରିନଥିଲୁ, କାରଣ:
- User ଯେ agent ଭିତରେ paste କରନ୍ତି ସେଠାରେ ପୂର୍ବରୁ codebase context ଅଛି — ସେମାନେ repo open କରି Cursor / Claude Desktop ବ୍ୟବହାର କରୁଛନ୍ତି।
- Per-(check × framework) templating କୌଣସି per-click cost ବିନା ~80% value cover କରେ।
- User ଚାହିଲେ ପରେ “ମୋ codebase ପାଇଁ AI ସହିତ refine କରନ୍ତୁ” opt-in API fire କରିପାରେ। ଆଜି, ନାହିଁ।
