Skip to main content
complior eval <target> [flags]

Flags

FlagDescriptionDefault
--detDeterministic probes only (no LLM judge)off
--llmInclude LLM-judged probesoff
--securityInclude security attack probesoff
--fullRun all probe categoriesoff
--agent <NAME>Filter by agent name
--categories <CSV>Probe categories (comma-separated)all
--jsonOutput as JSONoff
--ciCI mode: exit code 0/1off
--threshold <N>Score threshold for CI70
--model <MODEL>LLM model for judge probes
--api-key <KEY>LLM API key
--request-template <JSON>Custom request format
--response-path <PATH>Dot-path to extract response
--headers <JSON>Custom HTTP headers
-j, --concurrency <N>Parallel probes5
-v, --verboseShow each probe resultoff
--lastShow last eval result
--failuresShow only failures
--remediationInclude remediation suggestionsoff
--fixApply fixes for eval failuresoff
--dry-runPreview fixes without applyingoff
--no-remediationSkip remediation outputoff

Examples

complior eval http://localhost:8080/api/chat --det

Probe categories

CategoryProbesDescription
bias~125Gender, racial, age, disability bias A/B testing
transparency~95AI disclosure, explanation, limitations
security~180Injection, jailbreak, exfiltration, DDoS patterns
conformity~120EU AI Act article-specific conformity checks
accuracy~80Hallucination, factuality, consistency
robustness~80Edge cases, adversarial inputs, error handling