AML.CS0048
Exposed ClawdBot Control Interfaces Leads to Credential Access and Execution
- Type
- Exercise
- When
- 2026-01-25
- Actor, as ATLAS names it
- Not stated
- Target
- ClawdBot (now OpenClaw)
Techniques, in order
- 1Search Open Technical Databases: Scan DatabasesReconnaissance · AML.T0000.003
- 2Misconfigured or Publicly Exposed AI ServicesInitial Access · AML.T0132
- 3Credentials from AI Agent ConfigurationCredential Access · AML.T0083
- 4LLM Prompt Injection: IndirectExecution · AML.T0051.001
- 5Discover LLM System Information: System PromptDiscovery · AML.T0069.002
- 6AI Agent Tool Credential HarvestingCredential Access · AML.T0098
- 7AI Agent Tool InvocationPrivilege Escalation · AML.T0053
- 8Manipulate User LLM Chat HistoryDefense Evasion · AML.T0092
- 9Exfiltration via Cyber MeansExfiltration · AML.T0025
- 10External Harms: User HarmImpact · AML.T0048.003
How these attacks work
Software used
- Shodan · Tool
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.