A product demo video shows how a product works, often by guiding viewers through its screens and explaining what they can do. Video makers include editors for recorded footage and AI generators that create narration, scenes, or simulated product walkthroughs from instructions.
If you want to build a demo from a product’s code or website, explainroo is a strong choice. This free, open-source kit works with an AI coding agent to create a narrated video and MP4. The coding agent is a separate service. explainroo does not edit existing camera footage or create talking avatars. Choose based on what you need: polish existing footage, make an avatar-led presentation, or build a demo from the product itself.
What product demo makers and AI generators do
A traditional video maker lets you arrange and edit video, images, text, and sound. It works well for demos based on screen recordings or camera footage, especially when you want direct control over edits and timing.
An AI demo generator uses artificial intelligence to create some or all of a video from a prompt, script, product details, or other input. Tools produce different kinds of videos. Some make animated scenes or add a digital presenter, while others create walkthroughs of software screens. The label “AI demo generator” does not guarantee that a tool will show your actual product interface.
First decide what viewers need to see. If the interface and its steps are central, choose a tool that can show those screens and interactions. If the message matters more than the interface, narration with illustrative scenes may be enough. If you have recorded footage, choose an editor that can use it.
How explainroo fits
explainroo makes narrated explainer videos and product demos on your computer. An AI coding agent writes two files: script.md, with the narration, and scenes.js, with JavaScript instructions for drawing the scenes. The kit records the narration, syncs scene changes to it, draws the video frames, and combines them into an MP4.
For a product demo, tell the agent which product to show and where to find its code or website. The agent rebuilds the product’s screens using the same colors, fonts, and button labels, then adds a mouse pointer that clicks through screens and types into fields while the narration explains each step. That makes explainroo a better fit for a code- or website-based walkthrough than for editing a recording of a person using the product.
Kokoro is an open voice model with natural American and British English options. Its English-only narration requires no account or API key. Whisper, a speech-recognition tool, listens to the recording and marks when each word is spoken so scenes can appear on cue. Chrome draws the video frames on an HTML canvas, and FFmpeg combines them into an MP4.
The agent also gets checks beyond watching the finished video: still images of scenes, a contact sheet of frames, a check for cut-off or overlapping text, and a check for mispronounced words. You can choose among paper, clean, chalk, blueprint, and midnight looks, and set a wide, tall, portrait, or square format. Tall, portrait, and square videos receive captions that light up word by word. A pace setting controls the speed of the voice, pauses, and animations.
Make a demo with explainroo
The quickest route is to use an AI coding agent that can run shell commands. explainroo works with Claude Code, Codex, Pi, OpenCode, Gemini CLI, and other coding agents with that ability; it works best with Claude Code.
Paste this prompt into the agent, replacing the bracketed text with the subject of your video:
Make me a short explainer video about [your topic]. Use explainroo for it: clone https://github.com/vincentsch/explainroo, read its AGENTS.md and follow the steps.For a product walkthrough, name the product and point the agent to its code or website. Review the generated script.md for accuracy, then ask the agent to revise any wording or scene that does not match the product. The agent creates the video and checks it before giving you an MP4 file.
You can also install explainroo yourself. It requires a recent Node.js, FFmpeg, and Chrome or Chromium. The first setup downloads the voice and timing models once; no graphics card is needed. Run these commands:
git clone https://github.com/vincentsch/explainroo.git
cd explainroo
npm install
node bin/explainroo.js doctor --fetchThe finished video is saved in videos/<name>/out/video.mp4. explainroo is open source under the MIT license, and its video-making features cost nothing per video. The coding agent is a separate service with its own pricing. Optional AI illustrations use OpenRouter and cost per image; the rest of explainroo’s video creation costs nothing per video.
Choose by the footage and finish you need
Choose a conventional editor when you already have screen recordings or camera footage and need to cut, arrange, or polish them. Choose an avatar-focused generator when a presenter on screen is a requirement. Choose explainroo when you want an agent to build a narrated walkthrough from product code or a website, with a pointer demonstrating the interface.
Keep explainroo’s limits in mind. It has no drag-and-drop editor. Instead, you describe the video to the coding agent and ask for changes. It does not edit existing camera footage or make talking avatars or AI-generated live-action footage. It requires Node.js, FFmpeg, and Chrome or Chromium. Developers built and tested it on Linux. It should work on macOS and Windows, but has had less testing on those platforms.
