The Agent Skills Directory

[SAFE]: The skill provides educational material and code snippets following industry-standard practices for Next.js development. It includes patterns for data fetching, server-side rendering, and client-side interactivity without introducing malicious commands or hidden logic.
[INDIRECT_PROMPT_INJECTION]: The skill demonstrates patterns that ingest external data via search parameters, form submissions, and API request bodies. While these are standard web development features, they represent a theoretical attack surface for indirect injection if developers do not implement proper validation in their final applications.
Ingestion points: searchParams in app/products/page.tsx, formData in the checkout action, and request.json() in the API route handlers (references/details.md).
Boundary markers: The skill does not define agent-specific boundary markers as it is developer-facing documentation.
Capability inventory: The provided patterns demonstrate database interaction (db.product) and network requests (fetch).
Sanitization: The examples include basic presence checks for form data fields but rely on the developer to implement comprehensive input sanitization.

nextjs-app-router-patterns