Makinari Docs

Welcome to the Makinari documentation. This platform lets you deploy autonomous AI agents across multiple channels, driven by your business requirements.

Choose your path

Getting started depends on how you plan to use Makinari. Select the path that matches your goal:

For Business Operators

If you want to use the Makinari dashboard to set up agents and automate tasks:

1. Inbound Path: Add a tracking script, connect a chat widget or WhatsApp, and let agents handle incoming visitors.
2. Outbound Path: Import leads, build segments, and launch your first AI outbound campaign.
3. AI Goals: Give your agents brand context and assign them high-level Requirements to execute autonomously.

For Developers & Integrators

If you want to integrate Makinari into your own application or workflows:

1. First API Request: Create an API key, understand our dual authentication model, and make your first authenticated call.
2. MCP in 5 Minutes: Connect Cursor or Claude to Makinari’s Model Context Protocol server to give your local AI access to your project.
3. Webhooks & Events: React to Makinari events (like new leads or messages) in your own systems.

Core Concepts

Before diving deep, we recommend understanding the core concepts that power the platform:

• How Makinari fits together: Site → channels → agents → requirements → tools.
• Site and Members: The top-level workspace that owns all data.
• Agents and Commands: How our AI team operates under a token budget.
• Auth Model: When to use CORS vs. API keys.
• Environments and URLs: The canonical URLs for the dashboard, API, and legacy systems.
• Glossary: Standard terminology used across Makinari.