Guardrails

Kind: mechanism

Layer: Correctness Core

Record: lexicon:guardrails

Canonical: Lexicon

Constraints and filters that bound what a model is permitted to output or do at runtime.

Listed in Lexicon terms, after Grounding Strategy and before Input/Output Contract.

Category

• Model Architecture

Refactors

• Manual Runbook Dependency
• Artificial Intelligence Architecture
• Model Safety
• Agentic Architecture
• Auto-Remediation
• Canary Deployment

Required by

• Model Safety

Linked from

• Anti-patterns
• Model Architecture
• Observability / Auditability / Traceability
• Self-Healing / Recovery / Deployment Safety