Docs/Getting Started/Quick Start

Quick Start Guide

Get started with AgenticAnts and start monitoring your AI agents. Choose your preferred integration method below.

🎥

Want to learn more? Watch How to Set Up the ANTS Platform: AI Agent Tracing, Slack/Jira Integrations & Full LLMOps Walkthrough

Prerequisites

Before you begin, ensure you have:

  • A modern web browser
  • Node.js 20+ or Python 3.9+ installed
  • An existing AI agent or application (or follow along with our example)

Step 1: Create Your Account

  1. Visit agenticants.ai
  2. Click "Contact Sales" or "Get Started"
  3. Fill out the contact form with your details
  4. You'll receive an email with your account credentials
ℹ

Free Trial Available: Start with our free trial - 50,000 credits for 30 days, no credit card required.

Step 2: Set Up Your Organization and Project

After creating your account, you'll be guided through a quick setup process to configure your organization and project:

Organization and Project Setup

The setup workflow includes:

  1. Create Your Organization - Enter your organization name (e.g., "Acme Corporation")
  2. Invite Team Members - Add team members by email and assign roles (Admin, Member, Viewer). You can skip this and invite members later from Organization Settings
  3. Create Your First Project - Set up a project with a name and description. Select which team members should have access
  4. Add Payment Method - Optional step for continued usage beyond the free trial. Powered by Stripe for secure payment processing. Your free trial includes 50,000 credits for 30 days with no credit card required
  5. Setup Tracing - Get your API keys to start integrating AgenticAnts into your applications
ℹ

Setup Complete! Your organization and project are now ready. You can now proceed to get your API credentials.

Step 3: Get Your Credentials

Once logged in:

  1. Navigate to Settings → API Keys
  2. Find your Public Key and Secret Key
  3. Copy both keys and store them securely
bash
# Set your credentials as environment variables export ANTS_PLATFORM_PUBLIC_KEY="pk-ap-your-public-key-here" export ANTS_PLATFORM_SECRET_KEY="sk-ap-your-secret-key-here" export ANTS_PLATFORM_HOST="https://api.agenticants.ai" # Optional, defaults to production
⚠

Keep Your Keys Secret: Never commit your keys to version control or share them publicly. The secret key provides full access to your account.

Choose Your Integration Method

AgenticAnts offers two integration approaches. Choose based on your needs:

Auto-Instrumentation
2 minutes setup
  • ~10 lines of code
  • Automatic LLM tracing
  • Zero changes to existing code
  • Captures all LLM calls automatically
Full SDK
15 minutes setup
  • Full control over traces
  • Custom spans and metadata
  • Error handling patterns
  • You decide what to trace

Option A: Auto-Instrumentation (2 Minutes)

Use OpenLLMetry to automatically capture all LLM calls with zero code changes to your existing application.

ℹ

Supported Providers: OpenAI, Anthropic (Claude), AWS Bedrock, Azure OpenAI, Google Vertex AI, Gemini, Cohere, Mistral, Groq, LangChain, LlamaIndex, and more.

⚠

Language Support: Auto-instrumentation is available for Python and JavaScript/TypeScript only.

Step 4A: Install Dependencies

bash
pip install traceloop-sdk opentelemetry-exporter-otlp-proto-http

Step 5A: Initialize Auto-Instrumentation

Add this setup code once at the start of your application (before any LLM calls):

python
import os import base64 from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter from traceloop.sdk import Traceloop # Create exporter for AgenticAnts public_key = os.getenv('ANTS_PLATFORM_PUBLIC_KEY') secret_key = os.getenv('ANTS_PLATFORM_SECRET_KEY') auth = base64.b64encode(f"{public_key}:{secret_key}".encode()).decode() exporter = OTLPSpanExporter( endpoint="https://app.agenticants.ai/api/public/otel/v1/traces", headers={"Authorization": f"Basic {auth}"} ) # Initialize - this auto-instruments all LLM calls Traceloop.init(exporter=exporter)

Step 6A: Use Your LLM (No Changes Needed)

Your existing LLM code works as-is. All calls are automatically traced:

python
from openai import OpenAI # Your existing code - no changes needed! client = OpenAI() response = client.chat.completions.create( model="gpt-4", messages=[{"role": "user", "content": "Hello, world!"}] ) print(response.choices[0].message.content) # ✅ This call is automatically traced with tokens, cost, latency, and I/O

Now skip to Step 7: View Your Data to see your traces!

Option B: Full SDK (15 Minutes)

Use the AgenticAnts SDK for full control over your traces with custom spans, metadata, and error handling.

Step 4B: Install the SDK

Choose your preferred programming language:

bash
npm install ants-platform # or install specific packages npm install @antsplatform/tracing @antsplatform/otel # or yarn add ants-platform # or pnpm add ants-platform

Step 5B: Initialize AgenticAnts

Set up AgenticAnts in your application:

typescript
// For OpenTelemetry-based setup (recommended) import { NodeTracerProvider } from "@opentelemetry/sdk-trace-node"; import { resourceFromAttributes } from "@opentelemetry/resources"; import { SemanticResourceAttributes } from "@opentelemetry/semantic-conventions"; import { AntsPlatformSpanProcessor } from "@antsplatform/otel"; import { setAntsPlatformTracerProvider } from "@antsplatform/tracing"; // Create the Ants Platform span processor const processor = new AntsPlatformSpanProcessor({ publicKey: process.env.ANTS_PLATFORM_PUBLIC_KEY, secretKey: process.env.ANTS_PLATFORM_SECRET_KEY, baseUrl: process.env.ANTS_PLATFORM_HOST || "https://api.agenticants.ai", environment: process.env.ANTS_PLATFORM_ENVIRONMENT || "production", }); // Create tracer provider const provider = new NodeTracerProvider({ resource: resourceFromAttributes({ [SemanticResourceAttributes.SERVICE_NAME]: "my-agent", [SemanticResourceAttributes.SERVICE_VERSION]: "1.0.0", }), spanProcessors: [processor], }); // Register with Ants Platform and globally setAntsPlatformTracerProvider(provider); provider.register(); console.log('✅ Ants Platform tracing initialized');

Step 6B: Instrument Your First Agent

Now let's add monitoring to an AI agent:

typescript
import { startObservation } from '@antsplatform/tracing' import OpenAI from 'openai' const openai = new OpenAI() async function customerSupportAgent(query: string) { // Start a trace const trace = startObservation('customer-support-agent', { input: { user_input: query }, metadata: { agent: 'customer-support', channel: 'web-chat', priority: 'high' } }) // Set trace-level attributes trace.updateTrace({ userId: 'user_123', tags: ['agent:customer-support', 'service:my-app', 'version:1.0.0'] }) try { // Create a span for request validation const validationSpan = trace.startObservation('request_validation', { input: { raw_input: query }, metadata: { step: 'validation' } }) // Validate input... validationSpan.update({ output: { validated_input: query, input_length: query.length } }) validationSpan.end() // Create a span for LLM call const llmSpan = trace.startObservation('llm_preparation', { input: { model: 'gpt-4', task: 'customer-support' }, metadata: { step: 'preparation' } }) llmSpan.end() // Your agent logic const response = await openai.chat.completions.create({ model: 'gpt-4', messages: [ { role: 'system', content: 'You are a helpful customer support agent.' }, { role: 'user', content: query } ] }) const output = response.choices[0].message.content // Create generation span for LLM call const generation = trace.startObservation('gpt-4_llm_call', { model: 'gpt-4', input: query, output: output, usageDetails: { input: response.usage.prompt_tokens, output: response.usage.completion_tokens, total: response.usage.total_tokens }, metadata: { step: 'llm_call', status: 'success', agent: 'customer-support' } }, { asType: 'generation' }) generation.end() // Update trace with final results trace.update({ output: { result: output }, metadata: { input_tokens: response.usage.prompt_tokens, output_tokens: response.usage.completion_tokens, total_tokens: response.usage.total_tokens, status: 'success' } }) // End the trace trace.end() return output } catch (error) { // Log errors const errorSpan = trace.startObservation('error_handling', { input: { error: error.message }, output: { error_response: error.message }, level: 'ERROR', metadata: { error_type: error.name, step: 'error_handling' } }) errorSpan.end() trace.update({ level: 'ERROR', output: { error: error.message } }) trace.end() throw error } } // Use your agent const result = await customerSupportAgent('How do I track my order?') console.log(result)

Step 7: View Your Data

  1. Navigate to your AgenticAnts Dashboard
  2. You should see your first trace appear within seconds
  3. Click on the trace to see detailed information:
    • Input and output
    • Execution time
    • Token usage and cost
    • Metadata and tags

AgenticAnts Traces Dashboard

ℹ

** Congratulations!** You've successfully instrumented your first AI agent with AgenticAnts.

Next Steps

Now that you have basic monitoring set up, explore more features:

Advanced Tracing

Learn how to trace complex multi-agent systems with nested spans and parallel operations.

Read the SRE guide →

Cost Tracking

Set up detailed cost tracking and budgets for your AI operations.

Read the FinOps guide →

Security Setup

Add PII detection and security guardrails to protect sensitive data.

Read the Security Posture guide →

Framework Integrations

Integrate with LangChain, LlamaIndex, or other AI frameworks.

View all integrations →

Common Issues

Credentials Not Working

If you're getting authentication errors:

  1. Verify both your public key and secret key are correct
  2. Check they're properly set in your environment variables
  3. Ensure there are no extra spaces or characters
bash
# Verify your credentials are set echo $ANTS_PLATFORM_PUBLIC_KEY echo $ANTS_PLATFORM_SECRET_KEY

Connection Timeout

If requests are timing out:

  1. Check your network connection
  2. Verify firewall settings allow outbound HTTPS
  3. Try increasing timeout settings
typescript
const processor = new AntsPlatformSpanProcessor({ publicKey: process.env.ANTS_PLATFORM_PUBLIC_KEY, secretKey: process.env.ANTS_PLATFORM_SECRET_KEY, baseUrl: process.env.ANTS_PLATFORM_HOST, timeout: 30000, // 30 seconds })

Traces Not Appearing

If traces aren't showing in the dashboard:

  1. Verify your public key and secret key are correct
  2. Check the environment setting matches your dashboard view
  3. Verify the host URL is correct for your deployment
  4. Wait a few seconds - traces are processed asynchronously
  5. Check that your OpenTelemetry provider is properly registered
ℹ

Need Help? Join our Discord community or email support@agenticants.ai

Example Projects

Check out these complete example projects:

  • Customer Support Bot - LangChain + OpenAI integration
  • Code Assistant - Multi-agent system with tool usage
  • Document Q&A - RAG system with LlamaIndex
  • Data Analyst Agent - AutoGen integration with code execution

View examples on GitHub →

What's Next?

Learn Core Concepts → View Integrations →

© 2026 ANTS Platform, Inc.Docs v1.0 · Last updated June 2026