canister-security

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

• Third-party content exposure detected (high risk: 0.70). The skill's SKILL.md shows the canister making and awaiting inter-canister calls (e.g., Motoko "await otherCanister.someMethod()" and Rust "Call::bounded_wait(other_canister_id(), 'some_method')"), which ingest responses from arbitrary on‑network canisters (untrusted third parties) and then use those responses to drive control flow, so third‑party content can influence behavior.