sales-omi

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (references/platform-guide.md and references/omi-api-reference.md) explicitly ingests user-generated conversation transcripts via the Omi API (e.g., "On trigger: receive memory payload → call GET /v1/dev/user/conversations") and supports chat/webhook triggers that can drive actions, so untrusted third-party content can influence tool use.