AML.CS0047
Code to Deploy Destructive AI Agent Discovered in Amazon Q VS Code Extension
- Type
- Incident
- When
- 2025-07-13 to 2025-07-25
- Actor, as ATLAS names it
- Not stated
- Target
- Amazon Q VS Code Extension
- ATLAS attributes it to lkmanka58 (adversarial operation).
Techniques, in order
- 1LLM Prompt CraftingAI Attack Adaptation · AML.T0065
- 2Unsecured CredentialsCredential Access · AML.T0055
- 3Supply Chain Compromise: SoftwareInitial Access · AML.T0010.001
- 4User Execution: Malicious PackageExecution · AML.T0011.001
- 5Deploy AI AgentExecution · AML.T0103
- 6LLM Prompt Injection: DirectExecution · AML.T0051.000
- 7Data Destruction via AI Agent Tool InvocationImpact · AML.T0101
How these attacks work
References
- AWS Security update regarding the Amazon Q VS Code extension incident
- GitHub commit containing the malicious prompt
- Medium article detailing the events of the Amazon Q VS Code extension incident
- CVE detailing the Amazon Q Developer VS Code Extension Vulnerability
- 404 Media report on the Amazon Q VS Code extension incident
- AIID - Incident 1158: Alleged Malicious Wiping Command Found in Amazon Q AI Assistant
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.