codex-fleet-update

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime executes npm commands (npm view and npm install -g "@openai/codex@...") which fetch and install package code from the npm registry (https://registry.npmjs.org), causing remote code to be retrieved and executed on hosts.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill directs the agent to SSH into hosts, copy and install an executable across user accounts and recommends editing SSH config (RemoteForward) and proxy settings—actions that modify host state and SSH configuration and are security-relevant even though they don't request sudo or create users.