axiom-ai

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

• Third-party content exposure detected (high risk: 0.90). The skill's Tool protocol (see skills/foundation-models.md and skills/foundation-models-ref.md) explicitly describes calling external APIs and fetching data from third‑party sources (e.g., "Wikipedia API, news API", WeatherKit/MapKit, "any custom API") whose outputs are inserted into the LanguageModelSession transcript and used by the model to decide follow-up actions, so untrusted public content can materially influence tool calls and agent behavior.