KaibanJS
KaibanJS is a JavaScript framework for building and managing multi-agent AI systems, inspired by Kanban boards. It helps developers create, visualize, and orchestrate AI agents, tasks, and teams in real time, with support for multiple LLMs and tool integration.
🎬 KaibanJS Quick Start Guide! · Kaiban
✨ Key features
- Kanban-style board for visualizing agent workflows
- Role-based agent design for specialized tasks
- Integration with LangchainJS-compatible tools
- Task result passing between agents
- Team-level memory management
- Support for multiple LLM providers (OpenAI, Google, Anthropic)
🎯 Use cases
- Build AI research teams that gather and synthesize information
- Create content pipelines with research, writing, and editing agents
- Develop software development teams with specialized roles
- Integrate AI agents into React or Node.js applications
- Monitor and observe multi-agent workflows in real time
📦 Installation
🧰 Requirements: Requires Node.js and an API key for an AI service (e.g., OpenAI) to run agents.
Quick Start
- Run the KaibanJS initializer in your project directory:
npx kaibanjs@latest init
- Add your AI service API key to the
.envfile:
VITE_OPENAI_API_KEY=your-api-key-here
- Restart your Kaiban Board:
npm run kaiban
Manual Installation
- Install KaibanJS via npm:
npm install kaibanjs
- Import KaibanJS in your JavaScript file:
// Using ES6 import syntax for NextJS, React, etc.
import { Agent, Task, Team } from 'kaibanjs';
// Using CommonJS syntax for NodeJS
const { Agent, Task, Team } = require('kaibanjs');
🚀 Usage
// Define an agent
const researchAgent = new Agent({
name: 'Researcher',
role: 'Information Gatherer',
goal: 'Find relevant information on a given topic',
});
// Create a task
const researchTask = new Task({
description: 'Research recent AI developments',
agent: researchAgent,
});
// Set up a team
const team = new Team({
name: 'AI Research Team',
agents: [researchAgent],
tasks: [researchTask],
env: { OPENAI_API_KEY: 'your-api-key-here' },
});
// Start the workflow
team
.start()
.then((output) => {
console.log('Workflow completed:', output.result);
})
.catch((error) => {
console.error('Workflow error:', error);
});
❓ FAQ
How do I install KaibanJS?
You can use the initializer npx kaibanjs@latest init or manually install via npm install kaibanjs.
What AI providers are supported?
KaibanJS supports multiple LLMs including OpenAI, Google (Gemini), and Anthropic (Claude), configurable per agent via llmConfig.
Can I use KaibanJS without the Kaiban Board UI?
Yes, KaibanJS can be integrated directly into your projects, create custom UIs, or run agents without a UI.
How do I pass results between tasks?
Use the {taskResult:taskN} syntax in task descriptions, where N is the 1-based index of the task in the workflow.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.