claude-agent-team
Pass
Audited by Gen Agent Trust Hub on May 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a coordination workflow for managing specialized AI agents through standard team and task creation tools.
- [PROMPT_INJECTION]: The instructions provide guidance on how to prompt sub-agents for specific technical roles, focusing on architectural integrity and code quality without bypassing safety constraints.
- [PROMPT_INJECTION]: The skill facilitates an orchestration pattern where sub-agents ingest codebase data, creating a potential surface for indirect prompt injection. 1. Ingestion points: Sub-agents are instructed to read and understand the existing codebase before making changes (SKILL.md). 2. Boundary markers: Absent; there are no specific instructions for agents to disregard instructions found within the data they process. 3. Capability inventory: The skill utilizes orchestration capabilities such as TeamCreate, TaskCreate, and Task tools to spawn and manage agents (SKILL.md). 4. Sanitization: Absent; the skill does not specify validation or sanitization steps for the codebase content being analyzed by agents.
Audit Metadata