🚀 Socket Launch Week Day 5:Introducing Repository Access Permissions and Custom Roles.Learn more
Sign In

@useatlas/react

Package Overview
Dependencies
Maintainers
1
Versions
9
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

@useatlas/react

Embeddable Atlas chat UI for React applications

Source
npmnpm
Version
0.0.1
Version published
Weekly downloads
601
-30.84%
Maintainers
1
Weekly downloads
 
Created
Source

@useatlas/react

Embeddable Atlas chat UI for React applications.

Installation

npm install @useatlas/react

Usage

import { AtlasChat } from "@useatlas/react";
import "@useatlas/react/styles.css";

function App() {
  return <AtlasChat apiUrl="https://api.example.com" apiKey="your-key" />;
}

Custom Tool Renderers

Override how SQL results, charts, explore output, and Python results render inside the widget using the toolRenderers prop. This example assumes you have already imported the styles as shown above.

import { AtlasChat, type ToolRendererProps, type SQLToolResult } from "@useatlas/react";

function MySQLRenderer({ result, isLoading }: ToolRendererProps<SQLToolResult | null>) {
  if (isLoading || !result) return <div>Running query...</div>;
  if (!result.success) return <div>Query failed: {result.error}</div>;

  return (
    <div>
      <h3>Results ({result.rows.length} rows)</h3>
      <table>
        <thead>
          <tr>{result.columns.map((col) => <th key={col}>{col}</th>)}</tr>
        </thead>
        <tbody>
          {result.rows.map((row, i) => (
            <tr key={i}>
              {result.columns.map((col) => <td key={col}>{String(row[col] ?? "")}</td>)}
            </tr>
          ))}
        </tbody>
      </table>
    </div>
  );
}

function App() {
  return (
    <AtlasChat
      apiUrl="https://api.example.com"
      apiKey="your-key"
      toolRenderers={{
        executeSQL: MySQLRenderer,
      }}
    />
  );
}

Available tool names

Tool NameResult TypeDescription
executeSQLSQLToolResultSQL query results with columns and rows
exploreExploreToolResultSemantic layer exploration output (string)
executePythonPythonToolResultPython execution results with optional charts and tables

Custom renderers for any tool name are supported — just add the tool name as a key in the toolRenderers map. Tools without a custom renderer use the built-in defaults.

Renderer props

Every renderer receives ToolRendererProps<T>:

PropTypeDescription
toolNamestringName of the tool being rendered
argsRecord<string, unknown>Input arguments passed to the tool
resultTTool output. Built-in tool types include | null for the loading state
isLoadingbooleanWhether the tool invocation is in progress

Headless Hooks

For fully custom UIs, use the hooks entry point. Tool renderer types are also available here:

import { AtlasProvider, useAtlasChat } from "@useatlas/react/hooks";
import type { ToolRendererProps, SQLToolResult } from "@useatlas/react/hooks";

See the hooks documentation for details.

FAQs

Package last updated on 13 Mar 2026

Did you know?

Socket

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Install

Related posts