Skip to main content
IOC Radar

Indicators are what attackers leave behind. Stolen credentials are what they arrive with.

Check Exposure
1 IOC

Designing for the inevitable: System prompt leakage and mitigations in generative AI applications

AWS Security
AWS Security
Published July 8, 2026Original Report

Excerpt

System prompts form the foundation of generative AI applications. A system prompt is a collection of instructions and operational context provided to a large language model (LLM) that shapes how the model behaves and interacts with users and tools. System prompts often contain proprietary information, including role definitions, behavioral guidelines, tool descriptions and usage instructions, placeholders for conversation history and user metadata, Retrieval-Augmented…

Read the original report →

Attack Flow6 steps · MITRE ATT&CK mapped

Initial AccessTA0001T1190
1/6
Exploit Public-Facing Application
ActionCraft malicious inputs
Threat actors craft malicious inputs to manipulate generative AI applications.

Indicators of Compromise

Open the mindmap →

1 indicators

TypeIndicatorConfidenceScoreFirst Seen
SHA122d200f8670dbdb3e253a90eee5098477c95c23dHighScore 82/100First seen

IOC Relationship Graph

Building graph…