Introduction

This guide will walk you through everything you need to make your first request using the Brave Search API. From creating your account to calling the API, you’ll be up and running in just a few minutes.

Prerequisites

Before you begin, make sure you have:

• A valid email address for account registration
• A credit card to activate a plan
• Basic familiarity with making HTTP requests

Step 1: Create Your Account

Go to Brave Search API Dashboard to create your account:

1. Enter your email address and create a secure password
2. Verify your email address by clicking the confirmation link sent to your inbox
3. Log in and you are ready for Step 2.

Step 2: Activate a Plan

Once your account is created, you’ll need to activate a plan to access the API:

1. Navigate to the Plans page in your dashboard
2. Review the available plans and select one that fits your needs
3. Enter your credit card information

The Search plan includes both LLM Context (Search for AI agents) and Web Search.

Step 3: Create an API Key

After activating a plan, generate your API key:

1. Go to the API Keys section in your dashboard
2. Click on Add API Key
3. Give your key a descriptive name (e.g., “Production App” or “Development”)
4. Copy your API key and store it securely

Step 4: Make Your First Request

The Brave Search API uses a simple REST architecture. All requests require your API key in the X-Subscription-Token header.

LLM Context - Search for AI agents

Here’s how to retrieve pre-extracted web content for a model:

curl "https://api.search.brave.com/res/v1/llm/context?q=artificial+intelligence" \
  -H "X-Subscription-Token: YOUR_API_KEY"

import requests

url = "https://api.search.brave.com/res/v1/llm/context"
headers = {
    "Accept": "application/json",
    "Accept-Encoding": "gzip",
    "X-Subscription-Token": "YOUR_API_KEY"
}
params = {
    "q": "artificial intelligence"
}
response = requests.get(url, headers=headers, params=params)
results = response.json()

# Print the first extracted snippet
generic = results.get("grounding", {}).get("generic") or []
if generic and generic[0].get("snippets"):
    print(generic[0]["title"])
    print(generic[0]["url"])
    print(generic[0]["snippets"][0])

const url = new URL("https://api.search.brave.com/res/v1/llm/context");
url.searchParams.append("q", "artificial intelligence");

const response = await fetch(url, {
  headers: {
    Accept: "application/json",
    "Accept-Encoding": "gzip",
    "X-Subscription-Token": "YOUR_API_KEY",
  },
});
const results = await response.json();

// Print the first extracted snippet
const first = results.grounding?.generic?.[0];
if (first?.snippets?.length) {
  console.log(first.title);
  console.log(first.url);
  console.log(first.snippets[0]);
}

Understanding the Response

A successful LLM Context call returns extracted page content under grounding and source metadata under sources — ready to pass to a model:

{
  "grounding": {
    "generic": [
      {
        "url": "https://example.com/ai",
        "title": "Artificial Intelligence - Overview",
        "snippets": [
          "Relevant text chunk extracted from the page..."
        ]
      }
    ]
  },
  "sources": {
    "https://example.com/ai": {
      "title": "Artificial Intelligence - Overview",
      "hostname": "example.com"
    }
  }
}

Full documentation of responses at the LLM Context API Reference.

Building a search UI for humans? See Web Search instead.

Monitor Your Usage

Keep track of your API usage in the dashboard to:

• Stay within your plan limits
• Optimize your query patterns
• Plan for scaling needs

Next Steps

Congratulations! You’ve made your first search with the Brave Search API. Here’s what to explore next:

Need Help?

If you run into any issues or have questions:

• Check the LLM Context API Reference for parameters and schemas
• Review our Security Guidelines for best practices
• See the common questions and answers or contact our support team at Help & Feedback page