lean-agents

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions for task orchestration. It does not contain any executable scripts, binary files, or hidden commands.
  • [DATA_EXFILTRATION]: The instructions contain security-best practices, explicitly directing the agent to keep credentials out of task handoffs and use secure conversation channels instead.
  • [INDIRECT_PROMPT_INJECTION]: The skill addresses potential injection risks through architectural isolation, requiring exclusive file ownership for sub-agents and recommending isolated worktrees to prevent overlapping tasks from interfering with each other.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 02:14 PM