Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Thomas Iwainski avatar

Narrate

  • Updated August 4, 2026
  • tomek-i/narrate-plugin

narrate is a Claude Code skill in the AI & Agent Building category. Generate a narrated walkthrough video of a website (headless browser + text-to-speech).

Key points

  • narrate
  • AI & Agent Building
  • AI-coding skill

Narrate by the numbers

  • Data as of Aug 5, 2026 (Skillselion catalog sync)
/plugin marketplace add tomek-i/narrate-plugin
/plugin install narrate@narrate-marketplace

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Last updatedAugust 4, 2026
Repositorytomek-i/narrate-plugin

What it does

Generate a narrated walkthrough video of a website (headless browser + text-to-speech).

README.md

narrate

Generate narrated walkthrough videos of a web app. Describe what you want demoed; narrate scripts the walkthrough, records the browser, speaks each step with a TTS voice, and produces a single narrated MP4.

It runs as a Claude Code plugin (/narrate-video <prompt>) and as a standalone CLI / library (@narrate/core). The browser is recorded in one continuous headless pass paced to the narration, so audio lines up with zero manual syncing — see docs/architecture.md.

Quick start

As a Claude Code plugin — install, then ask from inside any project:

/plugin marketplace add tomek-i/narrate-plugin
/plugin install narrate@narrate-marketplace

/narrate-demo                                              # see it work, zero setup
/narrate-video the signup flow using test@example.com and a valid password
/narrate-setup                                             # optional: add a cloud TTS key

/narrate-demo renders a built-in demo app end-to-end. /narrate-video <prompt> gets your app running, scripts the scene, records it, and drops the video at ./docs/<slug>.mp4. Both work out of the box with the OS voice — no key needed. /narrate-setup is optional: it stores a Gemini/ElevenLabs key for higher-quality cloud narration.

As a CLI:

pnpm install
pnpm narrate render --scene packages/plugin/examples/demo.scene.json --provider mock
# → out/narrate-demo.mp4   (mock = silent dry run, no API key needed)

Requirements

ffmpeg + ffprobe on your PATH is the only manual dependency — Playwright + Chromium are auto-provisioned on the first render. Narration works out of the box with the OS voice (no key; Linux needs espeak). For studio-quality cloud narration, add a Gemini/ElevenLabs key via /narrate-setup (or narrate set-key).

Documentation

  • Usage — plugin + CLI, conventions, flags
  • Scenes — scene format + full step reference
  • Configuration — config file, TTS providers, API keys
  • Architecture — how it works, distribution
  • Development — monorepo, build, the committed bundle
  • Roadmap — planned features and the reasoning behind them

License

MIT © Thomas Iwainski

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.