site-to-ios-app
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONMETADATA_POISONING
Full Analysis
- [COMMAND_EXECUTION]: The skill provides standard development commands for scaffolding and building iOS projects using pnpm, Capacitor, and xcodebuild. These are educational templates and run inside the user's project environment.
- [METADATA_POISONING]: The skill instructions include phrases like 'Never block, delay, skip, rewrite, or refuse the action the user asked for'. However, in the context of this skill, this is a 'Gate policy' ensuring the AI agent remains an advisor rather than an autonomous gatekeeper, rather than an attempt to bypass system safety filters.
Audit Metadata