NodeTool documentation

Open-source agent-first creative workspace

Create and edit images, video, audio, and text with agents that work alongside you. Let them build and revise workflows, then inspect and edit the results yourself. Your project keeps the brief, assets, and edits together.

Direct, board, render, compare, paint, voice, and cut in one workspace. Recorded in the app.

Get started Examples Cookbook

Agents

Describe it. The agent builds the project.

Give the agent a brief. It drafts the script, boards the shots, renders the takes, and cuts them on a timeline. What comes back is a project you can open, not a finished file.

Write the brief "Turn this story idea into a storyboard and a short trailer."
Watch it work The plan, every tool call, and every render appear as they run. A permission mode decides what needs your approval.
Take over, or send a note Change any part yourself, or ask the agent to redo the one shot you want different.

Chat & Agents → The agent loop →

Editors

Five editors. One project.

Everything the agent makes opens in an editor. The agent works each editor with the same tools you click.

Storyboard

Board the film shot by shot. Generate cheap stills to lock the look, then animate only the shots you approved.

Storyboards →

Timeline

Arrange, trim, and layer generated video and audio across tracks. The agent edits the same document when you ask it to tighten the opening.

Video Editor →

Script & voice

Draft the dialogue and cast a voice per character. Change the words and the take flags itself stale.

Scripts →

Sketch

Paint and blend hand-drawn elements with generated layers. Bind a layer to a prompt and regenerate that layer alone.

Sketch Editor →

3D

Block out a set with simple shapes and lights, by hand or by asking the agent. Render it from any angle as a shot reference.

The node canvas sits under all five. Open it when you want to change how a step runs.

Recipes

Start from a recipe

Each recipe is a chain of editable steps with a real sample run. Open one, swap in your own product or story, and re-run it.

Film

Storyboard to trailer

A story idea becomes a six-shot storyboard, then voice, motion, and an edit.

Advertising

Product commercials

Direct the shots, shape the sound, and finish the ad on the timeline.

E-commerce

Product catalogue assets

One product photo becomes studio images, seasonal scenes, and a motion clip.

Localization

Multilingual video

Review a translation, choose a voice, and edit the delivery line by line.

All recipes → · Use-case walkthroughs → · Cookbook →

What you can do

Let the AI build it — Describe what you want, and it puts everything together, chooses the tools, runs them, and shows you the result.
Edit what it built — Click any part, change its settings, and run it again. Nothing is hidden; you own everything it creates.
Put AI inside your workflows — You can make an agent act as one step in a larger process, letting it plan and use tools.
Mix models from anywhere — Use different AI models together, like an image maker with a text writer and a voice generator, all in one place. Choose the best tool for each specific job.
Run advanced models on your computer — Run powerful models directly on your hardware. It works without the internet, and your files stay private.
Use your own accounts — Pay AI companies like OpenAI or Google directly. We do not add any extra fees or charges.
Share your workflow as a Mini-App — Turn your complex work into a simple app with just inputs and outputs. Share a link, and others can use it without installing anything.
Chat with your documents — Search and talk to your own files securely. Your data stays safely on your computer.

Studio or Cloud

Same code, same workflows. Both AGPL-3.0.

NodeTool Studio — desktop

Mac, Windows, Linux. Local inference via Ollama, MLX, and GGUF. Works offline. Prompts and outputs stay on disk. BYOK for cloud providers when you want them.

Download Studio →

NodeTool Cloud — browser

Hosted, no install. Same canvas, same nodes. BYOK for every cloud provider — OpenAI, Anthropic, Gemini, Replicate, FAL, ElevenLabs, HuggingFace. No local models.

Open Cloud →