- Home
- Skills
- Workflow Automation
- walkthrough video
More screenshots
Works with the AI Agents you already use
walkthrough video
Automated web app demo generator using Playwright and local AI voice to create narrated 1080p walkthrough videos.
$30
walkthrough video
Example session with this skill installed
Create a 2-minute walkthrough video for our CRM's new 'Contact Filtering' feature. Focus on the sidebar filters and the bulk export action. Use a professional tone.
- Read your context and instructions
- Compiled the walkthrough video
I have analyzed the CRM project structure and current Playwright tests. I have initialized the video-kit directory and configured the environment.
Storyboard: CRM Contact Filtering Walkthrough
| Scene | Visual Action | Narration |
|---|---|---|
| 1 | Title slide: CRM Filtering | Welcome to the new contact management experience. Today we will explore powerful filtering tools. |
| 2 | Navigate to /contacts | Start by navigating to the main contacts directory. The interface is clean and loads your full list. |
| 3 | Hover over sidebar | The new sidebar on the left contains all your segmenting options, including tags and status. |
| 4 | Select 'Lead' status | By selecting the Lead status, the list updates instantly. No page refreshes are required. |
| 5 | Toggle 'Active' tag | Combine filters to drill down. Here we are looking for active leads added this month. |
| 6 | Select all results | Once you have your segment, use the master checkbox to select every record in the view. |
| 7 | Click Bulk Export | The bulk export action is now accessible. This generates a CSV of your filtered segment. |
| 8 | Closing slide | Efficiently manage your pipeline with these new tools. Thanks for watching this tour. |
Next steps
- Review the narration text in the table above for tone and accuracy.
- Approve the scene sequence so I can generate the
chunks.txtfor the voice engine. - Ensure the test account has at least 10 'Active Lead' records for a realistic demo.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop relying on screen recorders that capture loading spinners and mouse stutters. This skill automates the production of high-quality web application walkthroughs by driving a real browser with Playwright and generating narration with local AI speech. It ensures your product demos are always accurate, reproducible, and professional without manual editing.
What it does
- Automated recording uses Playwright to capture app interactions while smoothing out cursor movements and removing wait times.
- Local narration generates high-quality voiceovers from text scripts using the Kokoro-ONNX engine.
- Smart assembly synchronizes video markers with narration timestamps to ensure the visual action matches the spoken word.
- Caption generation produces
.srtfiles and burns captions directly into the video for accessibility. - Dynamic overlays creates title cards, opening slides, and UI labels as part of the automated build process.
How it works
- Storyboard phase where you define the scenes, narration script, and specific app routes to be captured.
- Speech generation converts your script into localized WAV files and precise timing data for captions.
- Automated capture executes Playwright scripts to record clips, utilizing a specialized recorder for ripples and zooms.
- Final assembly processes all assets through FFmpeg to produce a 1080p MP4 with normalized audio.
Frameworks & tools
This toolkit uses Node.js and Playwright for browser automation, Python 3.11+ with Kokoro-ONNX for speech synthesis, and FFmpeg for video processing and encoding.
Why this beats prompting it yourself
Writing these scripts from scratch requires deep knowledge of FFmpeg filters and Playwright internal timing. This skill provides a pre-built orchestration layer that handles the complex math of retiming video clips to match variable speech lengths automatically.
Use cases
- Generate product demo videos that stay updated with your latest UI changes.
- Create onboarding tours for new users without hiring voice talent.
- Produce technical case studies or audits showing real app interactions.
- Build reproducible feature explainers for documentation sites.
Known limitations
Requires local installation of FFmpeg and Python audio libraries. Does not support capturing live microphone audio or system sounds during the recording process.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.
Requires Node 20+, Python 3.11+, and 350MB for local voice models. Cannot capture live system audio or mic. Requires FFmpeg with specific libraries like xfade and libx264.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago
- Software
Node.js
PythonPlaywright