跳转到主要内容
RAIL Score is built around five core primitives. Each has its own concept page, API reference, and SDK documentation.

Evaluation

Score content across 8 RAIL dimensions. The foundation of everything else.

Safe Regeneration

Automatically regenerate content that falls below your quality threshold.

Compliance

Check content against GDPR, HIPAA, EU AI Act, CCPA, and more.

Agent Evaluation

Intercept tool calls, scan results, and detect prompt injection in AI agents.

Middleware

Drop-in provider wrappers that score every LLM response automatically.

Policy Engine

Declarative rules that translate scores into block/warn/flag/allow actions.

How the pieces fit together

Start with Evaluation: every other concept builds on it.