skills/vercel-labs/emulate/next/Gen Agent Trust Hub

next

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • External Dependency Usage: The skill instructions involve installing and using packages from the @emulators scope, such as @emulators/adapter-next. These dependencies are used to provide the core emulation functionality within the Next.js environment.
  • Untrusted Data Ingestion Surface: The implementation of a catch-all route handler (app/emulate/[...path]/route.ts) creates a surface where the application ingests and processes external web traffic to emulate services like GitHub and Google.
  • Ingestion points: The catch-all route defined in app/emulate/[...path]/route.ts receives incoming requests to the configured paths.
  • Boundary markers: No specific delimiters or validation logic are shown in the configuration snippets, as the underlying library is designed to process standard web protocols.
  • Capability inventory: The skill is capable of rewriting response headers and HTML content, as well as persisting state to local files or external stores like Vercel KV.
  • Sanitization: Sanitization logic for the intercepted requests and rewritten responses is managed by the underlying library implementation rather than explicit skill instructions.
  • Local File Interaction: The optional persistence configuration using filePersistence involves writing emulator state to a local JSON file (.emulate/state.json). This is a typical pattern for maintaining state during development sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 10:00 AM
Security Audit — agent-trust-hub — next