Check Point Research
aidefend-sf0078
Claude Code Project Settings: Startup Execution and API-Key Theft
- Autonomous Agents
- AI Infrastructure
- System Compromise & RCE
- Credential & Identity Theft
- Data Exfiltration
Check Point researchers tested how Claude Code, an AI coding assistant, loads files from an untrusted repository. Separate hook, MCP startup and API-base-URL paths produced local command execution or API-key disclosure. The hook path required accepting repository trust; the other two could act before that decision. The demonstrated workspace-file access was limited to the tested key and workspace.
Mapped threat techniques
AML.T0081Modify AI Agent ConfigurationAML.T0050Command and Scripting InterpreterAML.T0091.000Application Access Token