orchestrating-swarms

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill consists of instructional content and configuration templates for platform-native multi-agent orchestration features.
  • [COMMAND_EXECUTION]: The skill describes the use of the Bash and Task tools to perform standard development operations such as git commands, file searches, and subagent spawning. These are intended functionalities of the platform being documented.
  • [EXTERNAL_DOWNLOADS]: Provides documentation for the installation of the it2 CLI utility from standard package registries (PyPI/NPM) to enable the iTerm2 visualization backend for agent processes.
  • [DATA_EXFILTRATION]: Documents the file-based messaging system used by agents located in ~/.claude/teams/ and ~/.claude/tasks/. These paths are part of the platform's internal state management and no unauthorized external exfiltration was observed.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 01:20 AM