Skip to main content

Build Intelligent AI Agents, Visually

Nadoo AI is an open-source AI Agent Builder Platform that lets you design, deploy, and manage production-ready AI agents through a visual workflow engine. Connect 12+ AI model providers, build RAG-powered knowledge bases, and deliver agents across Slack, Discord, Telegram, KakaoTalk, Teams, and WhatsApp — all from a single platform.

Why Nadoo AI?

Everything You Need in One Platform
  • Visual Workflow Engine: Design complex AI agent logic with a drag-and-drop graph editor
  • Knowledge Base with RAG: Ingest documents, build vector indexes, and power agents with retrieval-augmented generation
  • Multi-Channel Messaging: Deliver agents across 6+ messaging platforms without custom integration code
  • Chat Interface: Built-in chat UI with streaming responses, multimodal input, and conversation memory
Built for Real Workloads
  • Multi-tenancy: Workspace-based isolation with role-based access control (RBAC)
  • Async-first: FastAPI backend with async processing for high throughput
  • Background Tasks: Celery workers for document processing, embedding generation, and long-running operations
  • API Key Management: Secure credential storage for AI providers and external services
  • Docker Deployment: One-command setup with Docker Compose for local and production environments
Adapt and Extend
  • Plugin SDK: Python SDK for building custom tools that integrate with LLMs, knowledge bases, and storage
  • Skills System: Import and execute reusable skills from Git repositories
  • 18+ Node Types: AI Agent, Condition, Loop, Search Knowledge, HTTP Request, Code Executor, and more
  • 6 Agent Strategies: Standard, Chain of Thought, ReAct, Function Calling, Reflection, Tree of Thoughts
Community-Driven Development
  • Fully open-source under permissive license
  • Active community on GitHub and Discord
  • No vendor lock-in — self-host on your own infrastructure
  • Transparent roadmap and contribution process

Platform Components

Workflow Engine

The core of Nadoo AI — a LangGraph-based visual workflow engine where you connect nodes to build AI agent logic. Each workflow is a directed graph that processes user inputs through chains of AI reasoning, knowledge retrieval, conditional branching, and external tool calls.
Perfect for: Teams building complex AI agents that require multi-step reasoning, tool use, and conditional logic

Knowledge Base

Ingest documents (PDF, DOCX, TXT, Markdown) into a vector store powered by PostgreSQL with pgvector. The RAG pipeline supports hybrid search combining dense vector similarity with BM25 keyword matching, plus optional reranking for improved relevance.
Perfect for: Organizations that need AI agents grounded in their own documents and data

Multi-Channel Messaging

Connect your AI agents to the messaging platforms your users already use. Each channel integration handles authentication, message formatting, and bidirectional communication.
  • Slack — Bot integration with slash commands and thread support
  • Discord — Bot with channel and DM support
  • Telegram — Bot API integration
  • KakaoTalk — Channel and chatbot integration
  • Microsoft Teams — Bot Framework integration
  • WhatsApp — Business API integration

Knowledge Graphs

Build structured knowledge representations using RDF/OWL ontologies. Query with SPARQL and enable multi-hop reasoning across connected entities for complex question answering.

Plugin SDK and Skills

Extend the platform with custom Python plugins that access LLMs, knowledge bases, tools, and storage. The Skills system lets you import reusable executable units from Git repositories.

Architecture Overview

Quick Start Paths

Choose your path based on how you want to use Nadoo AI:
Get the full platform running locally with Docker.
1

Clone and Start

git clone https://github.com/nadoo/nadoo-ai-agent-builder.git
cd nadoo-ai-agent-builder
git submodule update --init --recursive
npm run start
2

Open the Platform

3

Build Your First Agent

Create a workspace, build a workflow, and test it in the chat interface.Follow the Quick Start guide

Use Cases

Customer Support Automation

Build AI agents that answer questions from your knowledge base, handle multi-turn conversations, and escalate to humans when needed — deployed across Slack, Teams, or your website.

Internal Knowledge Assistant

Ingest company documents into the Knowledge Base and let employees query policies, procedures, and technical documentation through natural language chat.

Multi-Channel Chatbot

Design a single AI workflow and deploy it simultaneously to Slack, Discord, Telegram, KakaoTalk, and WhatsApp with channel-specific formatting.

Document Processing Pipeline

Build workflows that ingest documents, extract structured data with AI, validate results, and route to downstream systems.

AI-Powered Data Analysis

Create agents that query databases, analyze data, generate visualizations, and produce reports using Code Executor and AI Agent nodes.

Workflow Automation

Chain together HTTP Request, Condition, and AI Agent nodes to automate complex business processes with intelligent decision-making.

Community and Support

Ready to Build?

Current Version: v0.5.0 — Nadoo AI Agent Builder Platform includes the visual workflow engine with 18+ node types, Knowledge Base with RAG, 12+ AI provider integrations, multi-channel messaging, Plugin SDK, and Skills system.

Built with care by the Nadoo AI team and contributors worldwide.