skills/raine/skills/workmux/Gen Agent Trust Hub

workmux

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation and usage examples for the workmux CLI. It instructs the agent on how to interact with the tool to manage isolated development environments.
  • [COMMAND_EXECUTION]: The skill describes various shell commands (e.g., workmux add, workmux list, tmux list-sessions). These are legitimate uses of development tools for managing git worktrees and terminal multiplexing. No malicious command injection or high-risk patterns were found.
  • [PROMPT_INJECTION]: While the skill uses instructional language like 'Do NOT explore' or 'you are a dispatcher', these are functional directions for task orchestration rather than attempts to bypass safety filters or override the system prompt.
  • [DATA_EXPOSURE]: The configuration section mentions copying .env files into worktrees. This is documented as a standard feature of the tool for managing isolated environments and does not constitute a secret leak or unauthorized access.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 10:38 AM
Security Audit — agent-trust-hub — workmux