agent-teams
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of instructional text and design patterns for multi-agent workflows. It provides conceptual guidance on team composition, sizing, and prompt engineering without implementing any functional logic.
- [NO_CODE]: There are no scripts, command-line executions, remote downloads, or tool invocations present in the skill files. The 'allowed-tools' field is empty, further restricting the agent's capabilities when using this skill.
- [DATA_EXPOSURE]: References to local documentation paths (e.g.,
~/.claude/docs/AGENT-TEAMS.md) point to the agent platform's internal help and reference system. These are not sensitive system files and do not involve data exfiltration or unauthorized access patterns.
Audit Metadata