cancel
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands like
rmandtmux kill-sessionto perform state cleanup and process termination. These operations are scoped to the project's.omcdirectory and the user's.claudeconfiguration directory, which is consistent with its intended purpose. - [DATA_EXFILTRATION]: The skill reads configuration from
~/.claude/teams/and identifies the repository root and remote URL to resolve state directories. This access is necessary for managing session-specific states, and no data is transmitted to external servers. - [REMOTE_CODE_EXECUTION]: The skill executes a local Node.js script and an embedded Python snippet for internal management tasks. These scripts are provided within the skill or plugin environment and are not fetched from remote sources.
Audit Metadata