spectacles-connected-lenses

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly shows reading and acting on publicly-writable, user-generated data (e.g., lensCloud.get('sharedNote','public_all') and RealtimeStore onValueChanged handlers) — both of which can contain arbitrary third-party content any lens/client can write and that the agent is expected to interpret and use to drive behavior.