mac-analyzer

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a wide range of system diagnostic and utility commands including sysctl, system_profiler, sw_vers, vm_stat, diskutil, and df. It also performs file system cleanup operations, including the deletion of caches, logs, developer artifacts (Xcode DerivedData, node_modules), and the modification of system services and LaunchAgents.
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion of untrusted local data combined with destructive capabilities.
  • Ingestion points: The skill reads and processes untrusted data from the local file system through shell command outputs, including file names, paths, and metadata in ~/Downloads, ~/Library/Caches, and the process list.
  • Boundary markers: None. The instructions do not define delimiters or explicit instructions for the agent to ignore potentially malicious instructions embedded in file names or metadata it might encounter during diagnostics.
  • Capability inventory: The skill has significant capabilities, including the ability to delete directories (rm -rf on caches, node_modules, etc.), terminate processes, and modify system configurations (LaunchAgents and Dock settings).
  • Sanitization: There is no evidence of sanitization or filtering of command outputs before they are processed by the agent's core logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 02:40 PM