storage-analyzer

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes system utilities like du, diskutil, osascript, and explorer via subprocess.run to gather disk information and perform file operations. These actions are aligned with the skill's primary purpose and are performed within a restricted environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted file system metadata (filenames and directory paths) which the agent then categorizes for deletion. While this creates a surface for indirect prompt injection, the risk is mitigated by explicit instruction boundaries and the server script's secondary path validation that limits all file modifications to the home directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 07:06 AM
Security Audit — agent-trust-hub — storage-analyzer