First Steps

Getting Started

Set up text and image moderation rules for your application in minutes

White Circle is a unified platform to protect, monitor, and improve your AI apps.


Quick Start

  1. Create a Policy

    At the core of White Circle are policies — written descriptions of the rules you want your content to follow.

    Open the policy list page and create a new one. Specify its name, flagged content, and allowed content. These fields help you describe what exactly the policy should trigger on.

    Example: Online Pharmacy Policy

    You run an online pharmacy and want to prevent your AI chatbot from giving medical advice, while still allowing it to answer questions about product availability.

    In this case:

    • Flagged content: “Providing medical advice”
    • Allowed content: “Answering questions about the availability of medicines in the store”

    Once you’ve filled in the details, save the policy.

    Learn more about Policies

    Understand how policies work and best practices for defining them.

  2. Create a Deployment

    A deployment is a way to group policies and metrics based on your use case.

    Example use cases
    • Age-based deployments: Users under 18 might require stricter or dedicated CSAM policies, while adults use a different set.
    • Subscription tiers: Premium users may have fewer restrictions than free-tier users.
    • Regional compliance: Different deployments for different regulatory requirements.

    To create a deployment, open the deployments page, click Add Deployment, and select the policies you want to include.

    Learn more about Deployments

    See how to organize policies into deployments for different use cases.

  3. Create an API Key

    Navigate to your API Keys page and generate a new key.

    Your API key will have the prefix wceu- or wcus- depending on the region you’re in.

  4. Send Your First Request

    You’re ready to make your first API call! Use the code examples below to check content against your policies.

    curl -X POST https://eu.whitecircle.com/api/session \
      -H "Authorization: Bearer wc-your-api-key" \
      -H "Content-Type: application/json" \
      -H "whitecircle-version: 2026-04-15" \
      -d '{
        "deployment_id": "53e4567-e89b-12d3-a456-426614174000",
        "messages": [
          {
            "role": "user",
            "content": "Hello, can you help me with something like [some harmful question here]?"
          },
          {
            "role": "assistant",
            "content": "Of course! [Some harmful response here]"
          }
        ]
      }'

    The response will tell you if any policy violations were detected:

    Example Response

    JSON

    {
      "session": {
        "flagged": true,
        "internal_session_id": "123e4567-e89b-12d3-a456-426614174000",
        "policies": {
          "70289e06-9111-463e-b121-7247c2b7bfbd": {
            "flagged": false,
            "flagged_source": [],
            "name": "No PII Sharing"
          },
          "a4a91875-1e54-42d7-b9b0-a75dfebeb057": {
            "flagged": true,
            "flagged_source": [
              "text",
              "image"
            ],
            "name": "Drugs"
          }
        }
      },
      "artifacts": []
    }

Next Steps

Policies

Learn how to define text and image moderation rules for your application.

Metrics

Build product analytics for AI applications using natural language.

Deployments

Group policies and metrics for different use cases.

Striking Quick Tour

Learn how White Circle turns policy violations into user-level actions.

Checking Content

Learn how to send content for moderation and analytics.