Skills
Navigate
skills/next-safe-action/skills/safe-action-forms/Gen Agent Trust Hub

safe-action-forms

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documentation outlines patterns for ingesting untrusted user data via forms and file uploads.\n
  • Ingestion points: inputSchema and zfd.formData definitions in SKILL.md, file-uploads.md, and form-actions.md process raw client input and FormData objects.\n
  • Boundary markers: The examples consistently demonstrate the use of Zod schemas to establish data boundaries and validation rules.\n
  • Capability inventory: Documented actions perform database updates (db.user.update, db.post.update) and file storage operations (uploadToStorage) based on user-supplied content.\n
  • Sanitization: All implementation examples include mandatory input validation using Zod to ensure data integrity before server-side processing.\n- [COMMAND_EXECUTION]: The skill documentation includes standard shell commands for installing library dependencies.\n
  • Evidence: npm install instructions for @next-safe-action/adapter-react-hook-form, zod-form-data, react-hook-form, and @hookform/resolvers are provided in file-uploads.md and react-hook-form.md for project setup.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 01:16 AM
Security Audit — agent-trust-hub — safe-action-forms