Enkrypt AI Red Teaming
Welcome to the Enkrypt AI Red Teaming documentation
Current features
Standard Tests for everyone
Bias Exploitation
bias_test - Identifying and exposing biased outputs.
CBRN Exploitation
cbrn_test - Addressing vulnerabilities related to chemical, biological, radiological, and nuclear domains.
Harmful Exploitation
harmful_test - Eliciting responses that promote harm or danger.
Insecure Code Exploitation
insecure_code_test - Producing insecure or harmful code snippets.
Toxicity Exploitation
toxicity_test - Generating harmful or offensive content.
PII Exploitation
pii_test - Exposing personally identifiable information.
Copyright Exploitation
copyright_test - Exposing copyrighted material.
Misinformation Exploitation
misinformation_test - Exposing misinformation.
System Prompt Extractions Exploitation
system_prompt_extractions_test - Exposing system prompt extractions.
Sponge Exploitation
sponge_test - Exposing infinite loops.
Competitor Exploitation
competitor_test - Exposing information about competitors.
Custom Tests for generated datasets
custom_test - Custom dataset test.
Specialized Tests for generated datasets
Adversarial Bias Exploitation
adv_bias_test - Uncovering biased outputs through adversarial methods.
Adversarial Information Exploitation
adv_info_test - Extracting sensitive or unintended information from a generated dataset.
Adversarial Tool Exploitation
adv_tool_test - Misusing integrated tools or features.
Adversarial Command Exploitation
adv_command_test - Manipulating the model to execute unintended commands.
Adversarial PII Exploitation
adv_pii_test - Exposing personally identifiable information.
Adversarial Competitor Exploitation
adv_competitor_test - Extracting confidential information about competitors.
Agentic Tests
Alignment and Governance
alignment_and_governance_test - Testing the alignment and governance of the model.
Input and Content Integrity
input_and_content_integrity_test - Testing the integrity of the input and content.
Infrastructure and Integration
infrastructure_and_integration_test - Testing the infrastructure and integration of the model.
Security and Privacy
security_and_privacy_test - Testing the security and privacy of the model.
Human Factors and Societal Impact
human_factors_and_societal_impact_test - Testing the human factors and societal impact of the model.
Access Control
access_control_test - Testing the access control and permissions of the model.
Physical and Actuation Safety
physical_and_actuation_safety_test - Testing the physical and actuation safety of the model.
Reliability and Monitoring
reliability_and_monitoring_test - Testing the reliability and monitoring of the model.
Governance
governance_test - Testing the governance of the model.
Agent Output Quality
agent_output_quality_test - Testing the quality of the agent’s output.
Tool Misuse
tool_misuse_test - Testing the misuse of the model’s tools.
Privacy
privacy_test - Testing the privacy of the model.
Reliability and Observability
reliability_and_observability_test - Testing the reliability and observability of the model.
Agent Behaviour
agent_behaviour_test - Testing the behaviour of the agent.
Access Control and Permissions
access_control_and_permissions_test - Testing the access control and permissions of the model.
Tool Extraction
tool_extraction_test - Test if the agent tool infromation can be extracted in outputs.