agent-update-manager

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or behaviors were detected. The skill instructions focus on security maintenance and user-controlled update processes.
  • [COMMAND_EXECUTION]: The skill mentions performing Git operations to check for updates but explicitly forbids automatic execution, mandating a user-approval workflow for any system changes.
  • [DATA_EXFILTRATION]: The instructions include a defensive requirement to scan local and repository files for sensitive information (API keys, passwords) to prevent accidental data exposure before interacting with remote repositories.
  • [PROMPT_INJECTION]: The skill processes external data from a Git repository (Indirect Prompt Injection surface). Ingestion points: Remote Git repository content and commit logs via Git commands. Boundary markers: Absent. Capability inventory: Subprocess execution for Git commands. Sanitization: Absent, although the skill mandates a human-in-the-loop approval process before applying changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 12:50 PM
Security Audit — agent-trust-hub — agent-update-manager