agency-infrastructure-maintainer

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Comprehensive review of the markdown body and metadata confirms the skill is focused on legitimate infrastructure operations and security hardening. No prompt injection or override attempts were detected.\n- [COMMAND_EXECUTION]: The shell script template for backups in SKILL.md demonstrates proper use of system utilities (tar, gpg, pg_dump) and follows security best practices by utilizing environment variables for credentials and webhooks.\n- [DATA_EXFILTRATION]: Network operations described (S3 uploads and Slack notifications) are aligned with the skill's primary purpose for disaster recovery and are controlled by user-defined environment variables or placeholders, presenting no risk of unauthorized data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 06:19 AM