pp-airframe

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

• Third-party content exposure detected (high risk: 0.70). This skill explicitly downloads and ingests open public third‑party datasets (FAA registry and the NTSB CAROL accident database via airframe-pp-cli sync/doctor) and uses those records to drive decisions (model safety, tail/event lookups), so untrusted external content can materially influence agent actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill instructs the operator to fetch and install remote binaries at runtime (e.g., via "go install github.com/mvanhorn/printing-press-library/library/developer-tools/airframe/cmd/airframe-pp-cli@latest" and "go install github.com/mvanhorn/printing-press-library/library/travel/flight-goat/cmd/flight-goat-pp-cli@latest"), which downloads and executes remote code and are required dependencies for the skill to run.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). Flagged because the skill instructs the agent to detect the OS and — with user consent — run system package-install commands that include sudo (apt/dnf) and other package managers, which modify system state and require elevated privileges.