- Home
- Skills
- Agents & Orchestration
- recording ui showcase videos
Works with the AI Agents you already use
recording ui showcase videos
Automate high-fidelity narrated web app demos with neural TTS and synchronized camera zooms.
Free
recording ui showcase videos
Example session with this skill installed
Create a 30-second narrated demo video for my local dashboard at http://localhost:3000 showing the user profile and settings page in dark mode.
- Read your context and instructions
- Compiled the recording ui showcase
I have generated the tour configuration and production script for your dashboard demo. The tour is configured for dark mode and uses speech-driven timing to ensure the narration perfectly matches the UI transitions.
Dashboard Demo Production
{
"name": "dashboard-walkthrough",
"out": "./recordings/dashboard-demo",
"colorScheme": "dark",
"timing": "speech",
"polish": {
"camera": true,
"frame": true,
"sound": true,
"cards": {
"title": "User Dashboard",
"subtitle": "Profile and Settings Walkthrough"
}
},
"steps": [
{ "goto": "http://localhost:3000", "hover": ".nav-profile" },
{ "caption": "Welcome to the user dashboard. Let's start by looking at your profile.", "focus": ".nav-profile" },
{ "click": ".nav-profile" },
{ "caption": "Here you can update your avatar and personal details.", "focus": ".profile-card" },
{ "click": ".settings-btn" },
{ "caption": "The settings menu allows you to manage notifications and security preferences.", "focus": ".settings-panel" },
{ "pause": 1000, "hold": true }
]
}
Next steps
- Run
npm installin the skill folder to ensure all dependencies and the TTS model are ready. - Execute
node scripts/record.ts path/to/tour.jsonto generate the raw recording. - Run
node scripts/tighten.ts ./recordings/dashboard-demoto remove silence and finalize the MP4.
Connects securely to your tools. The creator never sees your data.
About this skill
Manual screen recording for web app demos often results in jerky cursor movements, mismatched narration timing, and awkward silences during page loads. This skill automates the entire production pipeline, from driving a headless browser to generating neural text-to-speech that perfectly aligns with on-screen actions.
What it does
- Automated browser driving navigates Chrome using a JSON tour file to ensure consistent, repeatable walkthroughs.
- Neural narration generates high-quality voice-overs using Kokoro or Nabra, with support for multiple languages including Arabic.
- Smart timing alignment synchronizes spoken captions with on-screen events, eliminating manual editing for voice-to-video sync.
- Visual polishing adds automated camera zooms, motion blur, UI sound effects, and branded title cards without post-production software.
- Production tightening automatically removes dead air and speeds up non-essential transitions between spoken lines.
How it works
- Define the tour by creating a
<name>.tour.jsonfile specifying URLs, clicks, hovers, and the text to be spoken at each step. - Execute the recording using the provided TypeScript scripts to capture the browser screencast and generate the audio assets.
- Tighten the output to squeeze page loads and pointer glides into short gaps, ensuring the narration remains continuous and professional.
- Verify the quality using automated diagnostic tools that check for silence, volume levels, and frame-to-caption accuracy.
Frameworks & tools
The skill requires Node.js (v22.18+), ffmpeg, and Google Chrome. It utilizes the Kokoro neural TTS model for offline narration and supports optional network-based engines like Microsoft Edge TTS.
Why this beats prompting it yourself
Writing a script to record a browser is easy, but handling sub-second audio synchronization and dynamic camera zooming is difficult. This skill provides a specialized production pipeline that manages the complex relationship between speech duration and visual state transitions.
Use cases
- Generating product walkthrough videos for landing pages or documentation.
- Creating multilingual software demos without hiring voice actors.
- Producing high-fidelity bug reports that include narrated context and visible clicks.
- Scaling video content production for frequent UI updates and feature releases.
Known limitations
Requires a local installation of Node, Chrome, and ffmpeg. Pointer visibility is limited to simulated movements during tour steps. Latin words are dropped when using the Nabra Arabic engine.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Requires local Node.js, Chrome, and ffmpeg. Simulated pointer only moves on tour steps. Nabra engine drops Latin words. Narrated flow may drift on high-latency page loads.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 3 days ago
- Free with an Agensi account
- Accounts
GitHub
Opencollective
Feross
- Software
Python
Permissions
Allowed Hosts