Skip to Content
The Makinari API repo now includes an MCP Server — connect your AI models directly. View on GitHub →
AgentsAgents

Agents

Makinari provides a suite of specialized AI agents designed to handle specific business functions. Each agent is equipped with tools and capabilities tailored to its role.

How agents work

An agent is not a simple chatbot. It is a reasoning system that:

  1. Receives a goal (called a Requirement) with a token budget.
  2. Plans a sequence of actions to achieve that goal.
  3. Executes those actions using its available tools: sending messages, querying data, generating content, and more.
  4. Iterates based on results until the budget is exhausted or the goal is met.

The CMO agent acts as the strategic coordinator. When you assign a high-level Requirement, the CMO breaks it down into sub-tasks and delegates them to the appropriate specialized agents.

Agent setup

Before an agent can operate, it needs:

  • Channel access: The relevant channels must be configured (e.g., the Email agent needs a connected inbox).
  • Business context: Agents use your project’s brand information, tone guidelines, and knowledge base to stay on-brand.
  • Tool permissions: Each agent has a defined set of MCP tools it can call.
  • A goal: Agents do not act without an assigned Requirement or a direct trigger from the API.

Available agents

  • CMO - Chief Marketing Officer for strategy and analysis.
  • Sales - Lead generation and sales management.
  • Customer Support - Automated customer service.
  • Data Analyst - Data processing and insights.
  • Copywriter - Content creation and editing.
  • Email - Email management and automation.
  • WhatsApp - WhatsApp communication handling.
  • UX - User experience analysis.
  • Growth - Growth hacking and campaign management.
  • Chat - Real-time chat interactions.
Last updated on