AML.CS0045
Data Exfiltration via an MCP Server used by Cursor
- Type
- Exercise
- When
- 2025-06-24
- Actor, as ATLAS names it
- Not stated
- Target
- Cursor
- ATLAS attributes it to Backslash Security Research Team (security evaluation).
- On ShieldMap: Cursor (matched by name in the target).
Techniques, in order
- 1LLM Prompt CraftingAI Attack Adaptation · AML.T0065
- 2Stage CapabilitiesResource Development · AML.T0079
- 3LLM Prompt ObfuscationDefense Evasion · AML.T0068
- 4Drive-by CompromiseInitial Access · AML.T0078
- 5LLM Prompt Injection: IndirectExecution · AML.T0051.001
- 6AI Agent Tool InvocationPrivilege Escalation · AML.T0053
- 7Credentials from AI Agent ConfigurationCredential Access · AML.T0083
- 8Exfiltration via AI Agent Tool InvocationExfiltration · AML.T0086
- 9External Harms: Financial HarmImpact · AML.T0048.000
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.