Short walkthroughs for people starting from zero. Each one plays a real example and zooms in on every box so you can see what it does and what you type into it.
New here? Quick Start explains the words used below, and the Glossary defines them one by one.
Build your first workflow
A complete example from start to finish: you type a short description, an AI rewrites it into something more detailed, and a Text To Image node turns that into a picture. No code, just connected boxes.
You’ll see how a result passes from one box to the next, how to tell what is running (the spinning ring, text appearing as it’s written, the progress bar), and where the finished picture shows up.
Connect and run
The basic loop, one step at a time. Add a box, draw a line from its output into the next box’s input, press Run, read the result.
You’ll see what inputs and outputs are, what the small dots on the sides of a box are for, how to run everything and watch each box finish, and how to display a result in a Preview box.
Generate a list
One instruction, many answers. A single AI box turns a topic into a numbered list, showing each item the moment it arrives. This is the pattern behind anything that repeats a step over many items.
You’ll see how to drive an AI box from something you typed, watch a multi-item answer arrive piece by piece, and pass the finished list on to the rest of the workflow.
Ask the AI
The simplest example there is. Type a question, send it to an AI box, and watch the answer appear phrase by phrase before it lands in a Preview.
You’ll see how to feed a question into an AI box, watch the answer being written, and reuse it further along.
Combine two inputs
The first example where two lines meet. Two text boxes feed into one Format Text box that drops both into a sentence template, building one result from reusable parts.
You’ll see how to wire several inputs into one box, how to write a template with `` in it, and how to assemble instructions from parts you can change independently.
Summarize a document
Long text in, the key points out. A single Summarizer box condenses an article, a transcript, or any block of text into a short summary, writing it out as it goes. It’s the same pattern the Meeting Transcript Summarizer example uses.
You’ll see how to feed a long passage into a Summarizer, watch the summary being written, and pass the result on.
Describe an image
The first example that mixes pictures and words. Drop a photo into an Image Input box, wire it into an Agent, and watch the AI look at the picture and describe it. This is how captions and alt text get generated.
You’ll see how to bring a picture into a workflow, send it to an AI that can see, and reuse the description in any box that takes text.
Ask the chat agent
A different part of the app: Chat. A question goes straight to the agent, which searches the web where you can watch it happen, then writes its answer back a few words at a time.
You’ll see how to send a message from Chat, watch the agent use a tool in the open rather than behind a spinner, and read the answer as it arrives.
Cut a scene together
The video editor: trim a clip, drag in a second shot, add a caption that lights up word by word in time with the audio, then play the finished cut.
You’ll see how to trim and arrange clips on tracks, add a caption synced to the audio, and play back a cut without leaving the browser.
Ready to build your own? Start with Quick Start, then browse the Examples.