explainroo: narrated explainer videos made by your agent
explainroo
A skill and CLI: the agent writes a script and scenes, while the voice, word timing and MP4 render run on your computer with no API keys
- 793 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The tool installs dependencies through npm, downloads models and runs Chrome with scene code that the agent writes
- Demo mode captures screenshots, colors, fonts and wording from a site and reproduces the interface: someone else's brand or a misleading video draws complaints
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/explainrooDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The tool installs dependencies through npm, downloads models and runs Chrome with scene code that the agent writes
- Demo mode captures screenshots, colors, fonts and wording from a site and reproduces the interface: someone else's brand or a misleading video draws complaints
This is third-party code. Review the repository files before installing.
What it does
explainroo builds a video from two files the agent writes: `script.md` with the narration and `scenes.js`, a short drawing program. The open Kokoro model reads the script in eight languages, Japanese and Mandarin Chinese among them, and offers 54 voices. Whisper records which word is spoken when, and Chrome draws frames in the background, so a picture can appear on exactly the right word. Drawing supports charts, formulas, step-by-step code, your screenshots and video clips, and lines can look hand drawn. A camera follows the speech: it moves in on what the voice talks about and pulls back at the end so the last frame shows the whole idea. Background music and sound effects are synthesized for each video, with the music ducking under the voice. ffmpeg assembles the result: MP4, `.srt` and `.vtt` subtitle files, a cover image and YouTube chapters. Sizes fit YouTube, Shorts, TikTok, Reels and 4:5 or square posts. An agent cannot watch a video, so there are checks: scene snapshots, a contact sheet of small frames for the whole video, a search for cut off or overlapping text and a check of the spoken words. The `explainroo` skill tells the agent to find or install the tool and to read `AGENTS.md` with the scene rules.
Who it is for. For developers, product teams and educators who ask Claude Code or Codex for an explainer or product demo video and do not want to pay for cloud voice.
Good fit when
- You need a short video where a voice explains a topic, for example how HTTPS hides a password, and drawings appear as it speaks
- You need a product demo: the agent rebuilds the app's screens from its code or website and a mouse pointer walks through the scenario
- The video is for several platforms at once: one project renders to YouTube and TikTok sizes with `explainroo render`
- You want subtitles, a cover image and YouTube chapters along with the video, and a GIF for a README
- The narration must not go to the cloud: the voice and timing are computed on your computer
Not a fit when
- You need a live narrator with intonation or footage with actors: the scenes are drawn in code and synthetic voice stays synthetic
- The machine has no room for the voice and timing models, about 400 MB on first setup, or you cannot install ffmpeg and Chrome
- You plan to pass off an invented video as someone else's product or imitate another service's interface: demo mode reproduces colors, fonts and button labels
- You need a precise frame-by-frame storyboard made by hand: camera and animation follow the speech automatically
Example request
Make a 60 second video about how HTTPS keeps a password secret. Use explainroo: read its AGENTS.md and follow the steps, I want an MP4 at the endLimitations
You need Node.js 20.11 or newer, ffmpeg and Chrome or Chromium. First setup downloads the voice and timing models, about 400 MB, and no graphics card is needed. The author develops the tool on Linux, and tests run on Linux, macOS and Windows. Languages other than English need extra packages.
Every video carries a small "explainroo.com" mark in a corner. The `"watermark": false` setting in `video.json` turns it off, and the author asks you to keep it if you can. Illustrations through an OpenRouter image model are paid and sit outside the free part, and an ElevenLabs voice needs a separate account.
Claude Code, Codex, Pi and other agents all fit, and the tool works best with Claude Code and the newest Anthropic models. The cost of the agent itself and its terms are separate from explainroo.
How to disable. Delete the `~/.explainroo` folder holding the repository clone and the `explainroo` skill folder in the agent's skills directory. Finished videos live in `videos/<name>/out` inside your projects and stay in place.
FAQ
How do I install explainroo for Claude Code?
Paste the README prompt into the agent: clone `https://github.com/vincentsch/explainroo`, read `AGENTS.md` and follow the steps. By hand: `git clone https://github.com/vincentsch/explainroo.git`, then `npm install` and `node bin/explainroo.js doctor --fetch`. You can put the `skills/explainroo` folder into the agent's skills directory.
Do I need an API key for the voice?
No. The Kokoro voice and Whisper timing run on your computer, with no account or key. Keys are needed only for the optional parts: an ElevenLabs voice and illustrations through OpenRouter.
Which narration languages are supported?
English, Spanish, French, Italian, Brazilian Portuguese, Hindi, Japanese and Mandarin Chinese. European languages and Hindi need espeak-ng installed, Japanese and Chinese need Python with the misaki package.
How do I make a video for TikTok?
Set the size at render time: `explainroo render videos/demo --size tiktok` writes a vertical version next to the wide one. Vertical, square and 4:5 videos get captions that light up word by word.
Can I remove the watermark?
Technically yes: set `"watermark": false` in `video.json`. The author asks you to keep the "explainroo.com" mark in the corner, because that is how other people find the project.
Discussion
No questions or reviews yet.
Related
SkillsA Claude Code and Codex skill that turns a topic into a narrated motion-graphics explainer video on Remotion with subtitles and chapters
A skill for Cursor, Claude Code and Codex: from a product name or URL it builds a 4K promo with music, voice-over and checks
The video-explainer skill turns a script into a vertical 9:16 explainer: narration, captions, Remotion and a pre-render review
A skill for Claude Code and Codex that turns one sentence into a product launch film, scores the music and edits a frame on click
A skill that turns a coding agent into a studio for voiceover explainer videos: word-level sync, a motion library and Remotion assembly
A local video studio for your coding agent: describe a video in one sentence, the agent directs, your machine renders with no cloud
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| explainroo: narrated explainer videos made by your agent (this card) | no | 793 | |
| HyperFrames | no | 79,556 | builds MP4 files from HTML compositions the agent writes |
| anything2explainer | no | 3,088 | makes a narrated video on Remotion with subtitles and chapters |
README in short
The description follows the production stages: voice, timing, pictures, camera, sound and the final file. It then shows six looks, platform formats, pace settings and the product demo mode. Installation and credits to the libraries come last.
README badge
Original SKILL.md
---
name: explainroo
description: Make an explainer video (MP4) with a voice-over using explainroo. The voice is made on this computer and no API key is needed. Use when the user asks for an explainer, how-to, walkthrough, launch or tutorial video, an animated diagram with a voice-over, or a short for YouTube, Reels or TikTok.
---
# Make an explainer video with explainroo
explainroo makes an MP4 video from two files: a script with the words the
voice says, and a small JavaScript file that draws the pictures. A voice
model on this computer reads the script. explainroo writes down when each
word is spoken, Chrome draws the frames, and explainroo adds music and sound
effects.
## Find or install explainroo
Try these first:
```bash
command -v explainroo
ls "${EXPLAINROO_HOME:-$HOME/.explainroo}/bin/explainroo.js"
```
If neither exists, install it. It needs Node.js 20.11 or newer, ffmpeg, and
Chrome or Chromium.
```bash
git clone https://github.com/vincentsch/explainroo.git "$HOME/.explainroo"
cd "$HOME/.explainroo" && npm install
node bin/explainroo.js doctor --fetch
```
In the rest of this file, `explainroo` means whichever one you found: the
`explainroo` command, or `node "$HOME/.explainroo/bin/explainroo.js"`.
## Read AGENTS.md
`AGENTS.md` in the explainroo folder has the steps, the writing rules, the
settings and the Scene API, which is the list of everything `scenes.js` can
draw. Read it before you start.
## Make the video
Create the project where the user wants it, for example `./video`:
```bash
explainroo init video --theme paper --title "..."
```
Then follow the steps in AGENTS.md. When the video is done, tell the user
where `video/out/video.mp4` is, how long it is, and what
`explainroo verify video` said.