markusdowne/agentmail
Give AI agents their own email inboxes using the AgentMail API. Use when building email agents, sending/receiving emails programmatically, managing inboxes, handling attachments, organizing with labels, creating drafts for human approval, or setting up real-time notifications via webhooks/websockets. Supports multi-tenant isolation with pods.
95
93%
Does it follow best practices?
Impact
100%
1.07xAverage score across 4 eval scenarios
Low
Low-risk findings worth noting
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The SKILL.md and its references (references/webhooks.md and references/websockets.md) show the SDK ingesting inbound email bodies, attachments, webhook payloads, and websocket events (user-generated, untrusted content) and demonstrate agent code paths that read/process those messages and may trigger downstream actions (e.g., processEmail, auto-replies), so the agent is clearly exposed to untrusted third-party content that could enable indirect prompt injection.
- Audited
- Security analysis