updater

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool restricted to 'updater:*' commands. This grants the agent the ability to manage system software, including installation, deletion (moving to Trash), and configuration changes. These are intended functions for the software manager.
  • [EXTERNAL_DOWNLOADS]: The skill suggests installing the core CLI tool from a third-party Homebrew tap ('lu-zhengda/tap/updater'). While this is an external source, it is the primary vendor resource for this skill.
  • [REMOTE_CODE_EXECUTION]: The 'updater upgrade' and 'updater install' commands fetch and execute external packages or updates. This behavior is the tool's primary purpose and is restricted to the specific vendor utility.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection by ingesting untrusted application metadata.
  • Ingestion points: Metadata enters via 'updater check', 'updater scan', and 'updater history' commands (e.g., app names, release notes).
  • Boundary markers: No delimiters or safety instructions are used to wrap external content.
  • Capability inventory: The agent can execute any 'updater' command, including those that modify system files.
  • Sanitization: No sanitization of retrieved metadata is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 03:30 AM