Getting Started

This guide walks you through downloading HeartBeatAgents, configuring your first LLM provider, building an agent, and having your first conversation. The entire process takes less than five minutes.

1. Download & Install

Download HeartBeatAgents from www.heartbeatagents.com/download — it’s free, no account required. The free version includes 1 primary agent with full functionality, running entirely on your machine.

When you’re ready for more agents, choose a plan at www.heartbeatagents.com/pricing. After payment, we’ll send a magic link to your email to sign in and activate your license.

2. Add an LLM Provider

Agents need a language model to think. Navigate to Settings → Providers and add at least one provider. HeartBeatAgents supports OpenAI, Anthropic, Google, Mistral, Cohere, Meta, and DeepSeek.

To add a provider, you need an API key from that provider's platform. For example, to add OpenAI:

1. Go to platform.openai.com/api-keys
2. Create a new API key
3. Copy the key
4. In HeartBeatAgents, click "Add Provider"
5. Select "OpenAI" and paste your key
6. Click "Verify & Save"

The system validates your key by making a test API call. Once verified, the provider is available for all your agents. You can add multiple providers. HeartBeatAgents will route between them intelligently based on task requirements.

3. Create Your First Agent

Navigate to Agents → Create Agent. You will need to provide:

• Name: A display name for your agent (e.g., "Support Agent" or "Aria").
• Description: A brief summary of what this agent does.
• Primary Provider: Select the LLM provider this agent should use by default.
• Model: Choose a specific model from that provider (e.g., "gpt-4o" or "claude-3-sonnet").
• Standing Orders: The system prompt that defines your agent's role, personality, and boundaries.

Here is a minimal standing order to get started:

You are a helpful assistant for [Your Company].
You are friendly, concise, and professional.
If you don't know the answer to a question, say so honestly.
Never make up information.

Click "Create Agent" and your agent is live.

4. Start a Conversation

The fastest way to test your agent is through the built-in web console. On your agent's detail page, click "Open Console". This opens a real-time chat interface connected directly to your agent.

Type a message and press Enter. Your agent will process the message through the full pipeline: context loading, memory retrieval, LLM inference, and response generation. It will reply within a few seconds.

You can also test via the API:

curl -X POST https://api.heartbeatagents.com/v1/conversations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "agent_abc123",
    "messages": [
      { "role": "user", "content": "Hello! What can you do?" }
    ]
  }'

5. Connect a Channel (Optional)

To deploy your agent on Slack, Discord, or another platform, navigate to Channels → Connect Channel and follow the OAuth flow for your chosen platform. See the Channels Guide for detailed instructions per platform.

Next Steps

• Read Core Concepts to understand agents, memory, tools, and workflows.
• Connect integrations to give your agent access to external tools. See the Integrations Guide.
• Explore the API Reference for programmatic access to every platform feature.