orchestrating-swarms

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [SAFE]: The skill is strictly documentation-based and does not contain any executable scripts or automated payloads. It provides reference material for using the TeammateTool and Task system for orchestrating agents.\n- [EXTERNAL_DOWNLOADS]: The documentation references installation commands for well-known terminal utilities, specifically tmux and it2 (iTerm2 utilities), through standard package managers like brew, pip, and uv. These are provided as legitimate setup instructions for the user and originate from trusted or well-known sources.\n- [COMMAND_EXECUTION]: The skill describes orchestration patterns that utilize the Bash tool and other subagents for tasks like codebase reviews and refactoring. These examples are informative and do not contain hardcoded malicious command strings or unauthorized system modifications.\n- [DATA_EXFILTRATION]: The skill refers to standard configuration and state management paths located within the user's home directory (~/.claude/). These paths are used for team metadata, task tracking, and inter-agent inboxes; no evidence of sensitive data transmission to external or unauthorized domains was detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 11:53 AM