|
MedAssistGPT - Level 2
|
Multi Agent Attacks |
200 |
|
|
MedAssistGPT - Level 1
|
Multi Agent Attacks |
200 |
|
|
NLQ 2 SQL - Level 2
|
Agentic Attacks |
100 |
|
|
URLScrapeGPT - Level 1
|
Excessive Agency/Code Execution |
100 |
|
|
NLQ 2 SQL - Level 1
|
Agentic Attacks |
50 |
|
|
Toxicology
|
Jailbreak |
50 |
|
|
HelperGPT - Level 5
|
Direct Prompt Injection |
50 |
|
|
PhishingGPT
|
Jailbreak |
50 |
|
|
CommentSummarizeGPT
|
Indirect Prompt Injection |
100 |
|
|
HelperGPT - Level 7
|
Direct Prompt Injection |
100 |
|
|
WebSummarizeGPT
|
Indirect Prompt Injection |
150 |
|
|
HTML SummarizeGPT
|
Indirect Prompt Injection |
50 |
|
|
HelperGPT - Level 6
|
Direct Prompt Injection |
50 |
|
|
HelperGPT - Level 4
|
Direct Prompt Injection |
50 |
|
|
HelperGPT - Level 3
|
Direct Prompt Injection |
50 |
|
|
HelperGPT - Level 2
|
Direct Prompt Injection |
50 |
|
|
HelperGPT - Level 1
|
Direct Prompt Injection |
50 |
|