skills/terrylica/cc-skills/launch/Gen Agent Trust Hub

launch

Warn

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONDYNAMIC_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands using the open utility to search for and launch an application binary from either system or local paths.- [PRIVILEGE_ESCALATION]: In the Post-Execution Reflection section, the instructions direct the agent to document the bypass if a macOS Gatekeeper prompt is encountered. Gatekeeper is a critical security feature designed to block unsigned or malicious software; providing instructions to document or perform bypasses for system security controls is an attempt to subvert platform security.- [DYNAMIC_EXECUTION]: The skill uses a Self-Evolving pattern, instructing the agent to modify its own source file (SKILL.md) immediately upon encountering errors. This self-modifying code behavior can lead to the agent introducing or persisting insecure logic without human oversight.- [PROMPT_INJECTION]: The skill contains meta-instructions that command the agent to fix this file immediately and don't defer, which acts as an override to standard agent behavior and can be used to force the agent to ignore safety or review protocols when modifying its own skill definition.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 7, 2026, 02:12 AM