agentmail

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

• Third-party content exposure detected (high risk: 1.00). The skill accepts incoming emails via webhooks (see "Webhooks for Real-Time Processing" and the "Security: Webhook Allowlist" section in SKILL.md), which are untrusted, user-generated third-party messages that the agent is expected to read/interpret and which can directly influence actions—explicitly noted as a prompt-injection vector.