igrantio-api-holder

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

• Third-party content exposure detected (medium risk: 0.30). This skill’s holder runtime ingests outsider-authored free text via the notifications subsystem—GET /v2/config/digital-wallet/openid/notifications/sse streams notificationContent frames produced from external issuance/verification events.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs callers to fetch the raw specification at https://docs.igrant.io/openapispecifications/oid4vc.yaml at runtime to determine payload shapes, so external content can directly control the agent's request/behavior.