AI Posture Check Take the Check
AI security dimension

Prompt.

Injection testing, input validation, output filtering, OWASP LLM mapping, red-teaming.

Take the AI Posture Check All six dimensions
The six dimensions
Governance Data Prompt Model Runtime Vendor
Questions
2 in the adaptive tree
Frameworks
LLM01, MITRE ATLAS
Moves
The Controls axis of the Posture Quadrant
Why it matters

Why prompt matters

Prompt is where attackers meet your model. The Prompt dimension covers five interrelated controls: prompt-injection testing, input validation, output filtering, OWASP LLM Top 10 awareness and mapping, and red-teaming. Together they govern whether your AI deployments survive contact with adversarial input. OWASP LLM01 (Prompt Injection), LLM05 (Improper Output Handling), and LLM07 (System Prompt Leakage) all live here. The defining shift in this dimension is that classic application security practices (input validation, output encoding, threat modeling) need re-application against natural-language inputs and natural-language outputs. Most teams under-invest in this until an incident, then over-invest in tooling without process. The Posture Check helps separate signal from noise.

In the check

What the check asks about prompt

The check asks up to five questions in total, chosen by your earlier answers, so a path may meet one of these or none. Each carries a weight from 0 to 3 and cites its source.

QuestionCitationAnswers
Have your AI deployments been tested for prompt injection?Prompt injection is where written controls get tested first. OWASP LLM01
  • +0No testing
  • +1Informal manual testing
  • +3Formal testing in the release cycle
Do you red-team AI deployments before production, with a documented method?Audited controls get you to Run. Sprint means someone has tried to break them. OWASP LLM01, MITRE ATLAS
  • +0No red teaming
  • +1Informal red teaming
  • +3Formal red teaming, documented, on every release
Sources

Frameworks cited on this dimension

Every question above cites its source. These are the frameworks behind prompt.

Prompt Injection (LLM01)

An attacker manipulates an LLM through crafted inputs that override instructions, exfiltrate context, or trigger unintended actions. Direct .

Read the guide

MITRE ATLAS for AI Defenders

MITRE ATLAS (Adversarial Threat Landscape for AI Systems) is a knowledge base of tactics, techniques, and case studies for adversarial machi.

Read the guide
Other dimensions

The other five dimensions

Score yourself

See where prompt lands on your chart.

Five questions, about three minutes, and your dot lands on the Posture Quadrant. Your stage and the move that takes you forward arrive by email or in a live conversation with a CWS engineer.

Take the AI Posture Check Talk to a CWS engineer