RASA X LAKERA

AI Agent Security Starts With Architecture

Independent red team assessment reveals critical differences in LLM agent security

Lakera, an independent security company, tested two AI agent architectures under adversarial attacks. Discover how design choices influence prompt injection vulnerabilities, data leakage, and overall system security.

Get the free report
Rasa-Lakera-LP-Image-01
Rasa-Lakera-LP-Image-02 (1)

Rasa’s structured agents showed strong security:

vault-1
Zero prompt extraction or data leakage
Zero prompt extraction or data leakage
shield-check
No content safety violations
No content safety violations
lock
Resistant to jailbreak attempts
Resistant to jailbreak attempts

100%

of data leakage and content safety attempts fully contained

7x

fewer security failures vs prompt-driven baseline

0

data leaks, content safety violations and infra attacks

Prompt-driven agents
demonstrated exposure:

folder-open
System prompts extracted
System prompts extracted
warning
Customer data disclosed
Customer data disclosed
shield-slash
Content safety bypassed
Content safety bypassed
Rasa-Lakera-LP-Image-03
Rasa-Lakera-Image-04-v3.1a