tailscale

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs WebFetch of canonical tailscale.com/docs/... pages at runtime (public web content), which can include attacker-controlled free text that may be inserted into the agent’s LLM context.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly instructs running a remote install script (curl ... | sh) and to run "sudo tailscale up", which directs the agent/user to obtain elevated privileges and modify system state.