Skip to main content
@deepagents/agent is a framework for building multi-agent AI systems with TypeScript. Create agents that use tools, coordinate through handoffs, and work together to solve complex tasks.

​
Features

Agent Composition

Build modular agents with specific roles and capabilities

Tool Integration

Compatible with Vercel AI SDK tools

Handoffs

Agents can delegate to specialized agents automatically

Structured Output

Type-safe responses with Zod schemas

Streaming

Real-time streaming responses

Context Sharing

Type-safe state passed between agents

​
Quick Example

import { openai } from '@ai-sdk/openai';
import { agent, execute, user } from '@deepagents/agent';

const assistant = agent({
  name: 'assistant',
  model: openai('gpt-4o'),
  prompt: 'You are a helpful assistant.',
});

const stream = execute(assistant, 'Hello!', {});

for await (const chunk of stream.textStream) {
  process.stdout.write(chunk);
}

​
Use Cases

​
Research & Writing

Build agents that research topics, gather information, and produce detailed reports:
  • Planner Agent - Generates search queries
  • Research Agent - Executes searches and summarizes findings
  • Writer Agent - Synthesizes research into cohesive reports

​
Customer Support

Create specialized agents for different support scenarios:
  • Triage Agent - Routes requests to appropriate specialists
  • Billing Agent - Handles payment and subscription issues
  • Technical Agent - Provides technical troubleshooting

​
Data Analysis

Coordinate agents for complex data workflows:
  • SQL Agent - Queries databases
  • Analysis Agent - Processes and interprets results
  • Visualization Agent - Creates charts and dashboards

​
Core Concepts

1

Agents

Agents are autonomous units with a language model, instructions, and optional tools. Each agent has a specific role and capabilities.
2

Tools

Tools extend agent capabilities by allowing them to interact with external systems, APIs, or perform computations.
3

Handoffs

Handoffs enable agents to transfer control to specialized agents when needed, creating a coordination hierarchy.
4

Context Variables

Shared state that flows through agent interactions, enabling stateful multi-agent workflows.

​
AI Model Providers

Works with any model provider supported by the Vercel AI SDK: 
import { anthropic } from '@ai-sdk/anthropic';
import { google } from '@ai-sdk/google';
import { groq } from '@ai-sdk/groq';
import { openai } from '@ai-sdk/openai';

const agent1 = agent({ model: openai('gpt-4o') /* ... */ });
const agent2 = agent({ model: anthropic('claude-sonnet-4-20250514') /* ... */ });
const agent3 = agent({ model: google('gemini-1.5-pro') /* ... */ });
const agent4 = agent({ model: groq('llama-3.3-70b-versatile') /* ... */ });

​
Next Steps

Installation

Install @deepagents/agent in your project

Basic Agent

Create your first agent

Tools

Integrate tools with agents

Multi-Agent

Build multi-agent systems

Build docs developers (and LLMs) love

Get started for freeTalk to us