Dome’s configuration system lets you precisely control which guards run, how they execute, and what detectors they use. This guide covers all configuration options.
Configuration Hierarchy Dome organizes protection in three levels:
Guardrail (input/output) └── Guard (security, moderation, privacy) └── Detector (specific detection method)
Each level has its own configuration options that can be customized.
Guardrail Configuration Guardrail Options Option Type Default Description input-guardsList [] Guards to run on input output-guardsList [] Guards to run on output input-early-exitBoolean True Stop on first input flag input-run-parallelBoolean False Run input guards in parallel output-early-exitBoolean True Stop on first output flag output-run-parallelBoolean False Run output guards in parallel
Execution Modes
Early Exit (default) : Stops processing when the first guard flags content. Faster for rejecting clearly malicious input.
Complete Execution : Runs all guards regardless of flags. Useful for comprehensive logging.
Parallel Execution : Runs guards simultaneously for lower latency.Guard Configuration Guard Types Type Use Case Available Detectors securityAdversarial attacks Prompt injection, encoding detection moderationHarmful content Toxicity, profanity, hate speech privacySensitive data PII detection, secrets integrityData quality Format validation (experimental) genericCustom logic User-defined detectors
Guard Options Option Type Default Description typeString Required Guard category methodsList Required Detectors to use early-exitBoolean True Stop on first detector flag run-parallelBoolean False Run detectors in parallel blocked-responseString Default Custom block message
Detector Configuration Here are some of the available detectors:
Security Detectors: Detector Description Options prompt-injection-mbertMultilingual BERT model thresholdprompt-injection-deberta-v3-baseDeBERTa v3 model thresholdencoding-heuristicsBase64, Unicode tricks None security-embeddingsSemantic similarity threshold, top_ksecurity-llmLLM-based detection model_name
Moderation Detectors: Detector Description Options moderation-flashtextFast keyword matching wordlistmoderation-debertaNeural toxicity classifier thresholdmoderations-oai-apiOpenAI Moderation API None moderation-llamaguardLlama Guard model threshold
Privacy Detectors: Detector Description Options privacy-presidioPII entity recognition entities, thresholddetect-secretsCredential detection None
Work in Progress The programmatic protection capabilities and Dome integrations are currently in private preview and subject to change.
Next Steps
Use Guardrails Runtime integration patterns
Custom Detectors Build your own detectors
Observability Monitoring and tracing
Last modified on April 20, 2026
