Skip to content
explainroo
Video sizes and formats

How to Make a Video for YouTube with an AI Agent

Updated 3 min read

On this page
  1. Give the agent a clear video brief
  2. Make a video with explainroo
  3. Check the video before publishing

To make a YouTube video with an AI agent, give the software a goal and let it handle linked tasks such as writing narration and creating visuals. The agent can handle production steps. You still need to check the facts, choose what the video says, and approve the finished cut.

For narrated explainers and product demos, explainroo is the best fit: it is a free, open-source kit that lets a coding agent make a video on your computer. You describe the video, and the agent sets up explainroo, produces and checks the video, then gives you an MP4 file.

Give the agent a clear video brief

Start with a specific topic and viewer. “Explain how password managers work to someone who has never used one” tells the agent more than “make a technology video.” Tell the agent whether you want an explainer or product demo, what the viewer should learn, and which details or claims to check.

An AI agent differs from a chatbot that only drafts text: it can use tools to carry out several steps toward a goal. For a video, that might include writing narration, drawing scenes, and assembling the final file. Keep your own judgment in the process. Review factual claims and visual choices before publishing; YouTube’s monetization policies favor original, useful content over repetitive, mass-produced videos.

Make a video with explainroo

The quickest way to start is to use a coding agent that can run shell commands. explainroo works with Claude Code, Codex, Pi, OpenCode, Gemini CLI, and other coding agents; it works best with Claude Code.

Paste this prompt into the agent, replacing the bracketed topic:

Make me a short explainer video about [your topic]. Use explainroo for it: clone the explainroo repository, read its AGENTS.md and follow the steps.

The agent sets up explainroo, creates and checks the video, then gives you an MP4 file. It writes two files along the way: script.md, which contains the spoken words, and scenes.js, which describes how to draw each scene. You can time each scene to a specific word so the visuals match the narration.

explainroo uses the Kokoro open voice model to read the script in a choice of natural American or British English voices. Narration is English only, and you do not need an account or API key for the voice. Whisper, a speech-recognition tool, marks when each word is spoken to help time the visuals. Chrome draws the frames on an HTML canvas, and FFmpeg assembles them into an MP4.

You can choose among five looks in video.json: paper, clean, chalk, blueprint, and midnight. The kit also supports YouTube’s wide format, plus vertical Shorts, portrait, and square videos, with word-by-word captions for the latter three. To revise the video, ask your agent to change the script or scenes.

For a product demo, tell the agent the product name and provide its code or website. explainroo can rebuild screens using the product’s colors, fonts, and button labels, then show a pointer clicking and typing while the narration explains.

Check the video before publishing

An agent cannot watch the video it made. explainroo provides still images of scenes and a contact sheet of frames. It also checks for cut-off or overlapping text and mispronounced words. Review those checks yourself, then watch the exported video to confirm the narration, visuals, and pacing work together.

The kit makes its own background music and sound effects, and lowers the music while the voice speaks. YouTube lists AI-generated music among examples that require disclosure, so review the current disclosure guidance during upload. Also check that the title and thumbnail describe the actual video and that any outside assets have appropriate rights. YouTube notes that fair use depends on the circumstances and is ultimately decided by courts, not by an automated tool.

explainroo is for narrated explainers and product demos. It does not edit existing camera footage, make talking avatars, or generate live-action footage, and it has no drag-and-drop editor. The kit has no per-video cost. Optional AI illustrations through OpenRouter cost per image, and the coding agent is a separate service with its own pricing.

Free and open source

Let your AI agent make the video

explainroo lets a coding agent like Claude Code or Codex make narrated explainer videos and product demos. The voice, the word timing and the rendering run on your own computer, with no API key and no cost per video.

Copy this into your AI agent

Make me a 30 second vertical explainer video for TikTok about [your topic]. Use explainroo for it: clone https://github.com/vincentsch/explainroo, read its AGENTS.md and follow the steps.

Getting started Example videos GitHub

Made with explainroo: Why phone batteries wear out

More on video sizes and formats

All guides on video sizes and formats