Skip to content

LLM Chat

Built-in chat interface for interacting with LLMs directly.

Purpose

Use LLM Chat to:

  • Generate mock API responses
  • Test prompts before creating rules
  • Create realistic test data
  • Transform content with natural language

Requirements

Configure LLM settings in the Settings view first.

Interface

  • Multi-turn conversation with context
  • Token usage display per message
  • Provider indicator

Actions

Action Description
Send Submit your message
Save to Data Store Save assistant response for use in rules
Clear Reset conversation history

Use Cases

Generate Mock Response

Generate a realistic JSON response for a user profile API endpoint.
Include: id, name, email, created_at, subscription_tier

Then click Save to Data Store.

Test Prompt Variations

Try different prompt phrasings:

Explain how to implement authentication in Node.js

vs

As a senior developer, explain Node.js authentication best practices

Create Test Data

Generate 5 realistic error responses for a payment API:
- 400 Bad Request (missing fields)
- 401 Unauthorized
- 402 Payment Required
- 429 Rate Limited
- 500 Server Error

Transform Existing Data

Copy a response from Traffic view:

Transform this response to use fake PII:

{
  "user": {"name": "John Smith", "email": "john@example.com"}
}