bot-process-control

Warn

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to manage system processes and persistence via launchctl, pkill, and pgrep, including the management of launchd plist files in ~/Library/LaunchAgents/.
  • [CREDENTIALS_UNSAFE]: The skill provides instructions for interacting with sensitive OAuth token files in ~/.claude/tools/gmail-tokens/ and references secret storage within a 1Password vault.
  • [PROMPT_INJECTION]: The skill includes 'Self-Evolving' instructions that command the agent to modify the skill's own source code (SKILL.md) based on runtime performance, which can lead to unintended logic changes or behavioral drift.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 4, 2026, 09:52 AM