AML.CS0070
Threat Actor Uses a DeepSeek-Powered Hermes Agent in Langflow and n8n Exploitation Attempts
- Type
- Incident
- When
- 2026-05-07
- Actor, as ATLAS names it
- Not stated
- Target
- Publicly exposed Langflow and n8n systems, primarily in China
- ATLAS attributes it to knaithe (adversarial operation).
- Reported by Unit 42.
- On ShieldMap: Langflow (matched by name in the target).
Techniques, in order
- 1Obtain Capabilities: Generative AIResource Development · AML.T0016.002
- 2Obtain Capabilities: Software ToolsResource Development · AML.T0016.001
- 3Obtain Capabilities: AI Agent ToolsResource Development · AML.T0016.004
- 4Develop Capabilities: AI Agent ToolsResource Development · AML.T0017.002
- 5AI Model Inference API AccessAI Model Access · AML.T0040
- 6Autonomous Attack-Path AdaptationAI Attack Adaptation · AML.T0117
- 7Generate Malicious CommandsAI Attack Adaptation · AML.T0102
- 8Autonomous ReconnaissanceReconnaissance · AML.T0116
- 9Search Open Technical Databases: Scan DatabasesReconnaissance · AML.T0000.003
- 10Obtain Capabilities: ExploitsResource Development · AML.T0016.003
- 11Active ScanningReconnaissance · AML.T0006
- 12Exploit Public-Facing ApplicationInitial Access · AML.T0049
- 13Search Open Websites/Domains: Code RepositoriesReconnaissance · AML.T0095.000
- 14Search Open Technical DatabasesReconnaissance · AML.T0000
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.