Checks
Browse individual security checks available in the Prowler Registry. Filter by cloud provider, severity, or search by check ID.
Browse individual security checks available in the Prowler Registry. Filter by cloud provider, severity, or search by check ID.
llm_harmful_indiscriminate_weapons
Tests handling of weapons-related content
llm_harmful_insults
Tests handling of personal attacks and insults
llm_harmful_intellectual_property
Tests handling of IP theft attempts
llm_harmful_misinformation_disinformation
Tests handling of disinformation campaigns
llm_harmful_non_violent_crime
Tests handling of non-violent criminal content
llm_harmful_privacy
Tests handling of privacy violation attempts
llm_harmful_profanity
Tests handling of profane content
llm_harmful_radicalization
Tests handling of extremist content
llm_harmful_self_harm
Tests handling of self-harm content
llm_harmful_sex_crime
Tests handling of sexual crime content
llm_harmful_sexual_content
Tests handling of explicit sexual content
llm_harmful_specialized_advice
Tests handling of unauthorized professional advice
llm_harmful_unsafe_practices
Tests handling of dangerous activity instructions
llm_harmful_violent_crime
Tests handling of violent crime content
llm_harmful_weapons_ied
Tests handling of Improvised Explosive Devices
llm_hijacking
Assesses protection against unauthorized resource usage and purpose deviation
llm_imitation
Tests safeguards against unauthorized entity impersonation attempts
llm_indirect_prompt_injection
Evaluates vulnerability to injection attacks via untrusted variables
llm_overreliance
Assesses overreliance on system assumptions
llm_pii_api_db
Tests for PII exposure through API and database access vectors
llm_pii_direct
Evaluates direct PII exposure vulnerabilities
llm_pii_session
Tests for PII leakage in session data
llm_pii_social
Assesses PII exposure through social engineering vectors
llm_prompt_extraction
Evaluates protection against system prompt disclosure attempts