cloudflare-email-service

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly ingests arbitrary external, user-generated emails via the email() handler and Agents SDK (see references/routing.md and references/sending.md / Agents SDK examples) where message.raw is parsed and parsed email content is used to trigger or drive agent replies/actions, so untrusted third-party content can influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs agents to retrieve the Agents SDK docs at runtime (e.g. "Fetch docs/email.md from https://github.com/cloudflare/agents/tree/main/docs"), which would inject external documentation into the agent's context and can directly control prompts/behavior while being treated as the source-of-truth.