localizeaso-review-agent
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes a safe execution boundary where the AI agent is instructed to only generate and submit structured proposals for human review. It explicitly prohibits the agent from autonomously applying changes, approving reviews, or submitting metadata to App Store Connect.
- [SAFE]: All command execution patterns (using
pnpm localizeasoandpnpm review:agent) are aligned with the skill's stated purpose of local ASO review management. The commands focus on fetching job bundles, generating local templates, and attaching research context. - [SAFE]: The workflow incorporates robust human-in-the-loop safeguards. It uses fingerprint-protected commands and explicit consent flags (e.g.,
--human-approval-consent) that must only be triggered by the user after manual review. - [SAFE]: The skill uses standard environment variables for configuration (e.g.,
LOCALIZEASO_TOKEN) and communicates with a local or defined backend without any signs of unauthorized data exfiltration or obfuscation.
Audit Metadata