nextjs-expert

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references standard development dependencies including prisma, next-auth, zod, react-hook-form, and @tanstack/react-query for installation via npm. These are well-known community packages.
  • [COMMAND_EXECUTION]: In references/EXAMPLES.md, the skill includes standard shell commands for project scaffolding (npx create-next-app), package installation, and local development lifecycle tasks.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The BlogPost example in references/EXAMPLES.md fetches content from an external source through getPost(params.slug).
  • Boundary markers: The provided example (in references/EXAMPLES.md) does not include markers or instructions to isolate the rendered content from the agent's execution context.
  • Capability inventory: The skill environment defined in SKILL.md allows access to tools such as Bash, Write, and Edit.
  • Sanitization: The code example in references/EXAMPLES.md demonstrates the use of dangerouslySetInnerHTML to render post.content without explicit sanitization steps, representing a vulnerability surface for processing untrusted content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 05:13 AM
Security Audit — agent-trust-hub — nextjs-expert