next-partial-prefetching-adoption

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • Trusted External Tooling: The skill utilizes npx skills add to fetch auxiliary capabilities and npx @next/codemod to automate code transformations. These resources originate from the official Vercel/Next.js ecosystem, which is a recognized and trusted source for development tools.
  • Controlled Command Execution: The skill performs file system searches (rg, find) and executes project-specific build and test scripts (pnpm build, pnpm start, pnpm test). These actions are necessary for verifying the application's behavior during the migration process.
  • Testing API Configuration: The skill provides instructions for enabling exposeTestingApiInProductionBuild to allow end-to-end testing of prefetched content. It mitigates potential data exposure risks by recommending an environment variable gate (EXPOSE_TESTING_API === '1') and explicitly warning the user to avoid enabling this in public-facing production deployments.
  • Environment Sensitivity: The instructions guide the agent to identify and record necessary environment configurations and test credentials within a local instant-nav.rig.md file. This is a standard practice for configuring E2E testing environments and does not involve exfiltration to external services.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 12:37 AM