tunnel-doctor

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

• Third-party content exposure detected (high risk: 0.70). The skill's runtime workflow and tooling explicitly fetch and inspect arbitrary URLs and HTTP responses (see scripts/quick_diagnose.py which runs curl against a user-specified --url and the SKILL.md which shows numerous curl/diagnostic calls to arbitrary hosts and proxy APIs), so untrusted third‑party content can be ingested and influence diagnostic decisions and subsequent remedial actions.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The prompt includes explicit privileged operations (sudo snap remove, sudo tailscale up, installing packages via curl|sh) and instructions that modify system-level configuration, so it directs changes to the machine's state that require elevated privileges.