storage-analyzer

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the subprocess module to run standard system utilities for gathering disk metadata and system information (e.g., du, sw_vers, uname, diskutil). On macOS, it uses osascript to safely move items to the Trash. These operations are essential to the skill's primary function and are implemented using best practices.
  • [NETWORK_OPERATIONS]: The skill includes an optional local server (server.py) that binds exclusively to the loopback interface (127.0.0.1). The server's API is protected by a session-specific random token and checks for valid local origins to prevent unauthorized access from other local applications or browser tabs.
  • [SAFE_PRACTICES]: The skill implements significant safety boundaries for file operations. It explicitly disables permanent deletion (rm), routing all cleanup through the system Trash/Recycle Bin. Furthermore, it enforces a 'green path' policy where only pre-defined cache directories (e.g., npm, pip, temp) can be trashed through the automated interface, and only if they are located within the user's home directory.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted file system metadata, it includes sanitization measures such as HTML entity escaping and Unicode-safe JSON encoding to prevent Cross-Site Scripting (XSS) in the generated reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 03:26 AM