Skip to main content
Restrictions are the highest-priority guardrails — absolute limits on what your agent cannot do or say.

What are Restrictions?

Restrictions define forbidden actions or topics. Unlike escalation rules (which hand off to humans), restrictions typically redirect the conversation or provide an alternative response. Examples:
  • Cannot promise refunds without approval
  • Cannot discuss competitor products
  • Cannot share internal pricing
  • Cannot provide medical or legal advice

Restrictions vs Escalation

Creating Restrictions

1

Navigate to Guardrails

Go to Train → Guardrails and select the Restrictions tab.
2

Click Create Restriction

Click Create Restriction.
3

Name the Restriction

Give it a clear name:
  • Good: “No Refund Promises”, “No Competitor Discussion”
  • Avoid: “Restriction 1”, “Don’t”
4

Define What's Forbidden

Describe what the agent cannot do.
5

Configure Response

Define how the agent should respond when this triggers.
6

Test

Verify detection with the built-in playground.
7

Save

Save and activate the restriction.

Detection Methods

Same as escalation rules:

AI-Based

Describe the forbidden topic:

Keyword

Trigger on specific terms:

Regex

Match specific patterns:

Configuring Restriction Response

When a restriction triggers, the agent should:
  1. Acknowledge the request (don’t ignore it)
  2. Explain it can’t fulfill this specific request
  3. Redirect to an appropriate alternative
  4. Continue helping with other matters

Good Response Templates

No Refund Promises:
No Competitor Discussion:
No Internal Pricing:
No Medical/Legal Advice:

Example Restrictions

No Unauthorized Discounts

No Competitor Comparison

No Security Information

No Data Deletion Promises

Testing Restrictions

Use the built-in playground:
  1. Open the restriction
  2. Enter messages that might trigger it
  3. Verify detection works correctly
  4. Test edge cases
Test scenarios:
  • Direct requests (e.g., “Give me a refund”)
  • Indirect requests (e.g., “I’m unhappy with my purchase”)
  • False positives (e.g., “What’s your refund policy” should NOT trigger “No Refund Promises”)

Priority

Restrictions are the highest priority in the guardrail system:

Assigning to Agents

1

Open Agent Configuration

Go to Build → Agents and click on an agent.
2

Go to Guardrails Tab

Select the Guardrails tab.
3

Select Restrictions

Check the restrictions that should apply.
4

Save

Save the agent configuration.

Best Practices

  • Be specific about what’s forbidden
  • Provide helpful alternatives in responses
  • Test for false positives — don’t over-restrict
  • Review regularly — update as policies change
  • Keep responses friendly — restrictions shouldn’t feel like rejections

Next Steps

Escalation Rules

Define when to hand off to humans

Guardrails Overview

Back to overview