macdog

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the macdog command-line utility to perform system-level operations such as auditing security posture, managing firewall rules, and revoking TCC permissions. These actions are aligned with the skill's stated purpose of system hardening.
  • [EXTERNAL_DOWNLOADS]: It provides instructions to install the macdog tool via a Homebrew tap (lu-zhengda/tap/macdog). As this tap belongs to the skill author, it is considered a vendor-provided resource and is documented neutrally.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests system data. 1. Ingestion points: Output from macdog audit and macdog events (system logs). 2. Boundary markers: Absent. 3. Capability inventory: System configuration changes via macdog audit --fix and macdog firewall. 4. Sanitization: Not explicitly implemented in the prompt logic. However, given the local nature of the data sources, the risk is minimal.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 03:30 AM