agent-teams

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE] (SAFE): No malicious patterns, obfuscation, or sensitive data access detected.
  • [NO_CODE] (SAFE): This skill is strictly documentation and does not contain any executable scripts, binaries, or automated tasks.
  • [INDIRECT_PROMPT_INJECTION] (SAFE): While the orchestration patterns involve agents synthesizing data from sub-agents (a theoretical surface for indirect injection), this skill provides only static guidance and does not implement the data processing logic itself. * Ingestion points: SKILL.md (Patterns describe Lead synthesizing Specialist outputs). * Boundary markers: Encourages clear task descriptions but lacks technical delimiters. * Capability inventory: References management tools (TeamCreate, TaskCreate). * Sanitization: No sanitization logic present in the guidance.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:46 PM