AML.CS0067
Claude Code GitHub Action Secret Exposure
- Type
- Exercise
- When
- 2026-06-05
- Actor, as ATLAS names it
- Not stated
- Target
- Claude Code GitHub Action
- ATLAS attributes it to Microsoft Defender Security Research Team (security evaluation).
- On ShieldMap: Claude Code (matched by name in the target).
Techniques, in order
- 1Search Open Websites/Domains: Code RepositoriesReconnaissance · AML.T0095.000
- 2Discover AI Agent Configuration: Tool DefinitionsDiscovery · AML.T0084.001
- 3Discover AI Agent Configuration: Activation TriggersDiscovery · AML.T0084.002
- 4LLM Prompt CraftingAI Attack Adaptation · AML.T0065
- 5Prompt Infiltration via Public-Facing ApplicationInitial Access · AML.T0093
- 6LLM Prompt Injection: IndirectExecution · AML.T0051.001
- 7LLM JailbreakDefense Evasion · AML.T0054
- 8AI Agent Tool InvocationExecution · AML.T0053
- 9AI Agent Tool Credential HarvestingCredential Access · AML.T0098
- 10Data from AI Services: Conversation Context and MemoryCollection · AML.T0085.002
- 11Exfiltration via AI Agent Tool InvocationExfiltration · AML.T0086
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.