disk-cleaner

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute system commands such as du, df, and find to analyze disk usage across the user's home directory, including hidden directories and system library folders.
  • [COMMAND_EXECUTION]: It performs file system modifications, including changing file permissions via chmod -R u+w and deleting files during the interactive cleaning phase. Although it includes safety rules to avoid deleting source files or .git directories, the integrity of these operations depends on the agent's interpretation of command outputs.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data in the form of file names and paths from the local file system.
  • Ingestion points: Output from the find and du commands which traverse user-controlled directories (SKILL.md).
  • Boundary markers: Absent; there are no specific delimiters used to isolate system command output from the agent's instructional context.
  • Capability inventory: The skill has access to the Bash tool, enabling it to execute deletion and permission commands (SKILL.md).
  • Sanitization: None; the instructions do not specify any sanitization or validation for file paths before they are processed or used in deletion commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 07:17 AM