t3-remote-setup

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The workflow connects via SSH to an external remote host and runs commands whose output (such as system metadata, configuration files, and command outputs) is fed back into the agent context.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill executes a remote installation script downloaded via curl from https://t3.codes/install.sh, which represents an unverified third-party operational dependency.