mole-mac-cleaner

Fail

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONPRIVILEGE_ESCALATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The instructions explicitly recommend installing the 'Mole' tool by downloading a script from an external repository and piping it directly into a shell interpreter.
  • Evidence (SKILL.md): curl -fsSL https://raw.githubusercontent.com/tw93/mole/main/install.sh | bash
  • This pattern allows the remote script to execute arbitrary code with the user's current shell permissions.
  • [PRIVILEGE_ESCALATION]: The skill includes a command specifically for modifying system authentication settings to allow Touch ID to be used for sudo operations.
  • Evidence (SKILL.md): mo touchid # Configure Touch ID for sudo and mo touchid enable --dry-run
  • Modifying /etc/pam.d/sudo or equivalent system security files is a high-privilege operation.
  • [EXTERNAL_DOWNLOADS]: The skill relies on fetching content from an untrusted GitHub user account (tw93) at runtime.
  • Evidence (SKILL.md): https://raw.githubusercontent.com/tw93/mole/main/install.sh
  • [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing JSON output from the tool, which creates a potential surface for indirect injection if the system data being analyzed (e.g., file names or process info) is maliciously crafted.
  • Ingestion points: mo status --json and mo analyze --json outputs (SKILL.md).
  • Boundary markers: Absent; the prompt interpolation of JSON output lacks delimiters to separate data from instructions.
  • Capability inventory: The skill has extensive capabilities including file deletion (mo clean, mo purge, mo uninstall), network service resets (mo optimize), and privilege escalation (mo touchid).
  • Sanitization: Absent; the provided scripting examples (using jq) do not include validation or sanitization of the values before they are used in control flow or printed to the terminal.
  • [COMMAND_EXECUTION]: The primary function of the skill is to orchestrate a powerful CLI tool that performs destructive file operations, system optimization, and maintenance, providing a broad attack surface if the tool itself or the agent's interaction with it is compromised.
Recommendations
  • HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/tw93/mole/main/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 13, 2026, 03:16 PM