tfMITRE ATT&CK technique
Indirect Prompt Injection T1698
- Tactic
- Execution
- Platforms
- Windows, SaaS
- Version
- 1.0
- Created
- 18 September 2026
- Last modified
- 12 October 2026
- Contributors
- Girshel Chokhonelidze; Caio Silva
Adversaries may deliver malicious prompts to an LLM indirectly by embedding the prompts in content that the adversary expects may be processed by an LLM.[1] These prompts can be included in web pages,[2] documents, emails, tickets,[3] code, or other sources that an agent, copilot, or other AI system may retrieve and process. When an LLM processes this content, the embedded instructions may cause the system to perform unintended behaviors.[4]
ReconResource DevInitial AccessExecutionPersistencePriv EscStealthDef ImpairmentCred AccessDiscoveryLateral MoveCollectionC2ExfiltrationImpact
0
Threat groupsNone recorded
Threat groupsNone recorded · 27% of ATT&CK techniques have none either.0
CampaignsNone recorded
CampaignsNone recorded · 53% of ATT&CK techniques have none either.0
Tools & malwareNone recorded
Tools & malwareNone recorded · 30% of ATT&CK techniques have none either.0
Detection rulesNone recorded
Detection rulesNone recorded · 33% of ATT&CK techniques have none either.