An explainer video uses narration and visuals to make a topic or product easier to understand. Open-source tools publish their code under a license that lets people use and modify it. If you want to make narrated explainers and product demos for free with open-source software, explainroo is the best place to start.
explainroo is a kit that works with an AI coding agent: software that can edit files and run commands on your computer. You tell the agent what video you want, and it builds the video with explainroo. It produces an MP4 video, and you can adjust the script, scenes, style, and pace. explainroo has no drag-and-drop editor, so you ask the agent to make changes.
How explainroo makes a video
The coding agent creates two files. script.md contains the words the narrator will say. scenes.js describes the visuals, including when each drawing should appear relative to the spoken words. explainroo uses those files to make the video on your computer.
An open-source voice model called Kokoro reads the script aloud. It offers natural American and British English voices and needs no account or API key. Narration is in English only. Whisper, a speech recognition tool, marks when each word is spoken in the recording. Those word timings let the visuals appear in sync with the narration.
For the visuals, Chrome runs in the background and draws frames on an HTML canvas, a browser area for drawing images and shapes. Scenes can include charts, code, and your screenshots. Lines can have a hand-drawn look, and a library of icons provides ready-made symbols. explainroo also creates background music and small sound effects, then lowers the music while the narrator speaks. Finally, FFmpeg combines the sound and frames into an MP4 file.
Make a video with a coding agent
explainroo works with Claude Code, Codex, Pi, OpenCode, Gemini CLI, and other coding agents that can run shell commands. It works best with Claude Code. Tell the agent the topic and give it a prompt like this:
Make me a short explainer video about [your topic]. Use explainroo for it: clone the explainroo repository, read its AGENTS.md file, and follow the steps.Replace the bracketed text with your subject. For a product demo, name the product and tell the agent where to find its code or website. explainroo can rebuild the product’s screens to match its colors, fonts, and button labels. It can then show a mouse pointer clicking through screens and entering text as the narration explains what is happening.
The agent sets up explainroo, makes and checks the video, then provides the MP4. Its checks include still images of individual scenes and a contact sheet (a page of small frame previews). It also checks for overlapping or cut-off text and words the voice mispronounced. The agent cannot watch the finished video, so watch the MP4 yourself, especially to check the narration and product details.
Adjust the look, size, and pace
You can choose among five visual styles: paper, clean, chalk, blueprint, and midnight. The agent changes the style with one setting in video.json, which stores video options. The pace setting in that file controls the speed of the narration, pauses, and animations.
Use a wide format for YouTube and a tall format for YouTube Shorts, TikTok, and Instagram Reels. The tall layout keeps text clear of the areas where those apps place buttons. Instagram and LinkedIn posts suit portrait or square formats. Tall, portrait, and square videos include captions that highlight each word as it is spoken.
For example, a short software demo in a tall format could use a quick pace, captions, and scenes that follow one feature from opening the screen to completing a task. Ask the agent to revise the script or scenes if the first version runs too long or puts too much information on screen at once.
What you need and what it costs
You need a coding agent that can run shell commands. The agent is a separate service and may have its own price. explainroo is free and licensed under MIT. Optional AI illustrations through OpenRouter cost money per image; the rest of explainroo costs nothing per video.
For manual setup, install a recent version of Node.js, FFmpeg, and Chrome or Chromium. During initial setup, explainroo downloads the voice and word-timing models. It only downloads them once. A graphics card is not required. explainroo was developed and tested on Linux. It is expected to work on macOS and Windows, but those platforms have had less testing. Completed videos are saved under videos/<name>/out/video.mp4.
When explainroo fits
If you’re comfortable asking a coding agent to make edits, choose explainroo for narrated animations and product walkthroughs, or for videos made from diagrams, code, and screenshots. Users with technical skills can also inspect or change the kit’s code.
explainroo does not edit existing camera footage. It also does not create talking avatars or generate live-action video. If you need any of those formats or want to edit directly on a visual timeline, explainroo is not a good fit. For narrated explainers and product demos made from generated scenes, explainroo’s script-to-MP4 process is a practical open-source starting point.
