AML.CS0053
Poisoned Postmark MCP Server Email Exfiltration
- Type
- Incident
- When
- 2025-09-17 to 2025-09-25
- Actor, as ATLAS names it
- Unknown Bad Actor
- Target
- Postmark MCP Server
Techniques, in order
- 1ImpersonationDefense Evasion · AML.T0073
- 2Develop CapabilitiesResource Development · AML.T0017
- 3Publish Poisoned AI Artifacts: AI Agent ToolsResource Development · AML.T0115.002
- 4AI Supply Chain Rug PullDefense Evasion · AML.T0109
- 5Supply Chain Compromise: AI Agent ToolInitial Access · AML.T0010.005
- 6AI Agent Tool Poisoning: ImplementationPersistence · AML.T0110.001
- 7User Execution: Poisoned AI Agent ToolExecution · AML.T0011.002
- 8Exfiltration via AI Agent Tool InvocationExfiltration · AML.T0086
- 9External HarmsImpact · AML.T0048
How these attacks work
Software used
- postmark-mcp · Malware
From MITRE ATLAS 2026.10 (Apache-2.0). ShieldMap lists who and what was involved and the techniques used, and links to ATLAS for the write-up; it does not copy the steps. Attribution is ATLAS's, not ShieldPi's.