agentmail

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the agent to ask the human to "Paste the API key here" (and to insert it into a config), which requires the LLM to receive and handle a secret value verbatim and could lead to exfiltration.

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and displays incoming emails from external senders via the AgentMail API (see client.inboxes.messages.list in SKILL.md and scripts/check_inbox.py), which are untrusted, user-generated third-party content that the agent is expected to read and which could materially influence behavior (cron/webhook notification flows and inbox processing).