Email safety for AI agents: scan inbound for prompt-injection/phishing + outbound for secret leaks.
No remote MCP endpoint is published for this listing.
No capability manifest has been published for this listing yet.