Introduction

Radarkit tracks how AI assistants answer the questions your customers ask. For every answer it records whether your brand came up, where, in what tone, which websites the answer relied on and which competitors it named. This API gives you that data as JSON.

What you can get

You want to	Call
See how visible your brand is, day by day	Get visibility
Read the answers AI models gave, with what we found in them	List responses
Know which websites AI answers rely on	Top source domains
Compare your brand with competitors	Get share of voice
See what the model searched for behind the scenes	List query fanout
Pull thousands of rows into a file	Create export
Add prompts, or ask a prompt again right now	Add prompts
Have the Content agent write an article	Create article

Base URL

https://api.radarkit.ai/v1

Quick start

Create an API key

Open API keysClick Create an API key, then copy the key. It is shown once.

Check the key

Send it as a Bearer token. This call is free and tells you what the key can do.

curl https://api.radarkit.ai/v1/me \
  -H "Authorization: Bearer rk_live_YOUR_KEY"

import requests

response = requests.get(
    "https://api.radarkit.ai/v1/me",
    headers={"Authorization": "Bearer rk_live_YOUR_KEY"},
)
print(response.json()["data"]["credits"])

const response = await fetch("https://api.radarkit.ai/v1/me", {
  headers: { Authorization: "Bearer rk_live_YOUR_KEY" },
});
const { data } = await response.json();
console.log(data.credits);

Pick a project

Every project call needs the project id. List your projects and copy the id. This costs 1 credit.

curl https://api.radarkit.ai/v1/projects \
  -H "Authorization: Bearer rk_live_YOUR_KEY"

projects = requests.get(
    "https://api.radarkit.ai/v1/projects",
    headers={"Authorization": "Bearer rk_live_YOUR_KEY"},
).json()["data"]
project_id = projects[0]["id"]

const projects = await fetch("https://api.radarkit.ai/v1/projects", {
  headers: { Authorization: "Bearer rk_live_YOUR_KEY" },
}).then((r) => r.json());
const projectId = projects.data[0].id;

From here, any endpoint in the sidebar works. Try Get visibility for your brand’s score per day.

Every response has the same shape

Field	What it holds
`data`	The result. An object for one thing, a list for many.
`meta`	Facts about the request: the project, the date window, the models, paging.
`credits`	`charged` for this call and `remaining` on the account after it.
`request_id`	A unique id for this call. Quote it when you contact support.

Failed calls have an error instead of data. See Errors.

Dates

Give a range with from and to as YYYY-MM-DD. Both days are included. Leave to out for today. Leave from out for the window your dashboard shows. Day boundaries follow the project’s timezone. Every timestamp the API returns is UTC.

Paging

Long lists come in pages of up to 25 rows. Every page has has_more and next_cursor. Send next_cursor back as cursor with the same filters to get the next page. There is no total count. For the size of a data set, ask for a free export estimate.

Authentication

Who can create a key, what a key may do, how to keep it safe.

Credits and rate limits

What every call costs and how fast you can call.

MCP Documentation

Ask the same questions in plain language from Claude or Cursor.

Overview

Account

Projects

Topics

Prompts

Visibility

Responses

Sources

Competitors

Rankings

Query fanout

Web search

Sentiment

Exports

Content

Changelog

What you can get

Base URL

Quick start

Every response has the same shape

Dates

Paging

Next

Authentication

Credits and rate limits

MCP Documentation

​What you can get

​Base URL

​Quick start

​Every response has the same shape

​Dates

​Paging

​Next

Authentication

Credits and rate limits

MCP Documentation

What you can get

Base URL

Quick start

Every response has the same shape

Dates

Paging

Next