skills/freekmurze/dotfiles/livewire-4/Gen Agent Trust Hub

livewire-4

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides standard documentation for building Laravel Livewire 4 applications. No malicious patterns, obfuscation, or unauthorized access attempts were detected. The documentation focus on security practices like authorization and validation demonstrates safe intent.
  • [INDIRECT_PROMPT_INJECTION]: The skill templates handle user input via Livewire component properties. While this creates an input surface, the skill instructs the agent on proper security measures. 1. Ingestion points: Public properties in SKILL.md and references/properties.md. 2. Boundary markers: Explicit security warnings in all reference files. 3. Capability inventory: Component rendering and database operations. 4. Sanitization: Recommends Laravel's authorization and validation systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:26 PM