NodeTool: Node-Based Visual Builder for AI Workflows and LLM Agents

Build AI Workflows. Run Them Locally.

NodeTool is a node-based visual programming tool for building AI workflows and applications. Connect models and tools with visual nodes to create LLM agents, RAG systems, and multimodal pipelines. Runs locally on macOS, Windows, and Linux — use local models or cloud APIs. Your data stays on your machine.

Key Features

What You Can Build

• AI Agents & Automation — multi-step agents that plan, execute, and adapt
• Document Intelligence — index documents, search with AI, and answer questions (RAG made simple)
• Image & Video Creation — generate and transform media with FLUX, NanoBanana, and custom models
• Data Processing — transform data, extract insights, and automate reports
• Voice & Audio — transcribe, analyze, and generate speech with Whisper and ElevenLabs
• Smart Assistants — AI assistants that understand documents, emails, and notes
• Mini-Apps — share workflows as interactive web applications

Cloud Models

Access the latest generative AI models through simple nodes:

Use TextToVideo, ImageToVideo, or TextToImage nodes and select your provider and model.

Some models need direct API keys. Others work through kie.ai, which combines multiple providers and often has better prices.

How NodeTool Differs

vs ComfyUI: ComfyUI focuses on Stable Diffusion image generation. NodeTool covers the rest of the AI stack: LLMs, RAG, audio, and video.

vs n8n: n8n automates business processes and APIs. NodeTool is built for AI work, with model management and local LLMs included.

vs LangChain: LangChain is a Python framework for LLM apps. NodeTool is a visual, TypeScript-first platform with an async Node.js runtime and custom nodes in TypeScript or Python.

Download

Flatpak CI Builds are also available for Linux.

Cloud-only usage requires no GPU — just use API services.

Documentation

• Getting Started — Build your first workflow
• Node Packs — Available operations and integrations
• Custom Nodes — Extend NodeTool
• Deployment — Share your work
• API Reference — Programmatic access

CLI & Server (npm)

Use NodeTool headless — run the server, execute workflows, or chat with agents from the terminal:

# Install globally (Node.js 24+ required)
npm install -g @nodetool-ai/cli

# Start the API server (port 7777)
nodetool serve

# Interactive AI chat with agent mode
nodetool-chat --agent --provider anthropic --model claude-sonnet-4-6

# Run a TypeScript DSL workflow
nodetool workflows run my-workflow.ts

# One-off without global install
npx --package=@nodetool-ai/cli nodetool serve
npx --package=@nodetool-ai/cli nodetool-chat --agent

See the CLI Reference for all commands.

Architecture

NodeTool is a monorepo with a TypeScript backend, React frontend, Electron desktop shell, and React Native mobile app.

nodetool/
├── packages/          # Backend monorepo (28 packages)
│   ├── kernel/        #   DAG orchestration & workflow runner
│   ├── node-sdk/      #   BaseNode class & node registry
│   ├── base-nodes/    #   100+ built-in node types
│   ├── agents/        #   Agent system with task planning & tools
│   ├── runtime/       #   Processing context & LLM providers
│   ├── websocket/     #   HTTP + WebSocket server (entry point)
│   ├── vectorstore/   #   SQLite-vec vector database
│   ├── code-runners/  #   Sandboxed code execution
│   └── ...            #   Protocol, config, auth, storage, deploy, etc.
├── web/               # React frontend (Vite + MUI + React Flow)
├── electron/          # Electron desktop app
├── mobile/            # React Native mobile app (Expo)
└── docs/              # Jekyll documentation site

For a detailed architecture overview, see ARCHITECTURE.md.

Development Setup

Prerequisites: Node.js 24.x, npm. Python 3.11 with conda for Python nodes (optional).

Node 24 is required. Electron 39 embeds Node 24 — native modules must match. Use nvm use to activate the correct version (reads .nvmrc).

Quick Start

nvm use                    # Activate Node 24 (reads .nvmrc)
npm install
npm run build:packages     # Build all TS packages in dependency order

# Run backend (port 7777) and frontend (port 3000)
# Uses tsx --watch for the backend, so startup skips a full websocket package rebuild.
npm run dev

Python Nodes (optional)

Python nodes (HuggingFace, MLX, Apple integrations) run via the PythonStdioBridge, which spawns a Python worker process that communicates over stdin/stdout. The bridge connects lazily on the first workflow that uses Python nodes — no separate setup is needed for the TypeScript backend.

Electron App

npm run electron

The Electron app auto-detects your active Conda environment. Settings are stored in:

• Linux/macOS: ~/.config/nodetool/settings.yaml
• Windows: %APPDATA%\nodetool\settings.yaml

Mobile App

cd mobile && npm install && npm start

See mobile/README.md for full setup.

Make Commands

Testing

# Unit tests
cd electron && npm test && npm run lint
cd web && npm test && npm run lint

# Web E2E (needs backend on port 7777)
cd web && npx playwright install chromium && npm run test:e2e

# Electron E2E (requires xvfb on Linux headless)
cd electron && npm run vite:build && npx tsc
cd electron && npx playwright install chromium && npm run test:e2e

For detailed testing documentation, see web/TESTING.md.

Contributing

We welcome bug reports, feature requests, code contributions, and new node creation.

Please open an issue before starting major work so we can coordinate.

License

AGPL-3.0

Get in Touch

• General: [email protected]
• Team: [email protected], [email protected]

GitHub | Discord | Website

Star History

🚀 Quick Start

# Install globally (Node.js 24+ required)
npm install -g @nodetool-ai/cli

# Start the API server (port 7777)
nodetool serve

# Interactive AI chat with agent mode
nodetool-chat --agent --provider anthropic --model claude-sonnet-4-6

# Run a TypeScript DSL workflow
nodetool workflows run my-workflow.ts

# One-off without global install
npx --package=@nodetool-ai/cli nodetool serve
npx --package=@nodetool-ai/cli nodetool-chat --agent