creating-subagents

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides purely educational content and templates for subagent configuration without including any executable scripts or malicious instructions.
  • [SAFE]: The documentation actively promotes security best practices, such as using the principle of least privilege by explicitly defining tool sets and recommending selective Bash access (e.g., Bash(git:*)).
  • [SAFE]: No evidence of prompt injection, data exfiltration, or obfuscation was found. The instructions for 'Proactive Invocation' are standard configuration guidance for the platform's delegation logic.
  • [SAFE]: The skill correctly identifies and warns about the security implications of MCP tool inheritance when tool restrictions are omitted, guiding users toward safer configurations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 07:36 PM