Skip to main content

Authorize (Phase 2)

The Authorize phase defines what the agent is allowed to perform. Configure guardrails, policies, and behavioral rules to enforce governance.

Access via Agent Detail → Authorize tab.

Authorization Pipeline

Operations flow through three layers:

Choosing the Right Layer

Each layer solves a different class of problem. Use the table below to decide which layer fits your use case.

LayerReach for this when…Example
GuardrailsYou need to validate or transform data flowing in/out — content safety, PII, banned termsMask credit-card numbers before they reach the LLM
PoliciesYou need a stateless permission check on a single operation — field-level conditions, thresholds, role gatesBlock invoice creation above $1,000 without approval
Behavioral RulesYou need to detect multi-step patterns across a session — sequences, frequencies, combinationsHalt file generation if the agent never queried the database

How Multiple Rules Execute

Guardrails, Policies, and Behavioral Rules can all have multiple rules active at the same time. The key difference is how they execute.

Guardrails run all enabled guardrails in order, like a pipeline. The output of one guardrail feeds into the next, which allows chaining transformations.

Input → Guardrail 1 (mask PII) → Guardrail 2 (mask bad words) → Guardrail 3 (block harmful content) → Output

Policies execute based on the logic defined in your Rego file. Multiple rules can exist within a single policy.

Behavioral Rules are checked one by one in priority order and stop at the first rule that triggers a verdict. Remaining rules are not evaluated.

Rule 1 (not triggered) → Rule 2 (triggered → REQUIRE_APPROVAL) → STOP — Rule 3, 4, 5... are skipped.

FeatureMultiple active?Execution
GuardrailsYesRuns all in order (chained)
PoliciesYesExecutes based on Rego logic
Behavioral RulesYesStops at first triggered verdict

Governance Decisions

The authorization pipeline produces one of four decisions:

DecisionEffectTrust Impact
HALTTerminates entire agent sessionSignificant negative
BLOCKAction rejected, agent continuesNegative
REQUIRE_APPROVALPauses for HITLNeutral (pending)
ALLOWOperation proceedsPositive (compliance)

Trust Tier-Based Defaults

Lower trust tiers receive stricter defaults:

TierDefault Behavior
Tier 1Most operations allowed, logging only
Tier 2Standard policies enforced
Tier 3Enhanced checks, some HITL
Tier 4Strict controls, frequent HITL

Next Phase

Once you've configured governance controls:

Monitor — Start your agent and observe its runtime behavior with Session Replay