Introduction

​

What is SpinAI?

SpinAI is a lightweight TypeScript framework for building AI agents that can:

• Execute actions in a specific order
• Maintain state between actions
• Make decisions using any LLM
• Handle complex dependencies

​

Key Concepts

​

Agents

Agents are the core of SpinAI. They:

• Receive user input
• Decide which actions to take
• Execute actions in the correct order
• Provide responses back to users

const agent = createAgent({
  instructions: "Help users with support tickets",
  actions: [getCustomerInfo, createTicket],
  llm: createOpenAILLM({ apiKey: process.env.OPENAI_API_KEY }),
});

​

Actions

Actions are discrete tasks that your agent can perform:

• Have clear inputs and outputs
• Maintain state through context
• Can depend on other actions
• Are executed in DAG order

const getCustomerInfo = createAction({
  id: "getCustomerInfo",
  description: "Retrieves customer information",
  async run(context) {
    context.state.customer = await fetchCustomer();
    return context;
  },
});

​

Context

Context flows through your actions:

• Preserves state between actions
• Carries the original user input
• Accumulates results as actions run

​

Why SpinAI?

• 🎯 Focused: Built specifically for action-oriented AI agents
• 🔄 Flexible: Works with any LLM that supports chat completions
• 📦 Simple: Minimal boilerplate, maximum productivity
• 🛠️ Type-safe: Built with TypeScript for robust development
• 🪶 Light: Zero dependencies beyond your chosen LLM

​

Next Steps