make-bot-ui

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

• Third-party content exposure detected (medium risk: 0.30). The skill instructs the agent to host a web UI reachable via Tailscale, where external peers can send JSON payloads that the agent later consumes as chat input.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the execution of an unpinned installation script fetched over HTTPS from tailscale.com (https://tailscale.com/install.sh), which constitutes an unverifiable runtime external dependency from an official publisher evaluated at high risk due to arbitrary remote script execution.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill prompt instructs the agent to run installation and setup commands using sudo (curl ... | sudo sh and sudo tailscale up), which violates security boundaries regarding privilege escalation.