| name | setup-pva-notebooklm |
| description | Overview of the Panda Video Automation pipeline — from NotebookLM research to video upload. Run this skill to see the full workflow. |
Panda Video Automation — Pipeline Overview
This project automates the full pipeline: research a topic via NotebookLM → generate AI video → prepare for upload → publish to platforms.
Prerequisite: Setup
npm install -g @panda-video-automation/pva
pva --version
Also ensure notebooklm-py is installed:
pip install notebooklm-py
notebooklm skill install
Pipeline Steps
Each step below resolves immediately (background work). Wait times are noted.
Step 1: Research a Topic
Creates a NotebookLM notebook and starts deep web research.
notebooklm create "<topic>" --json
notebooklm use <notebook-id>
notebooklm source add-research "<query>" --mode deep --no-wait
Wait ~5 min — run notebooklm research wait --import-all to finish and auto-import all sources.
Step 2: Generate Video
Generates a video overview in Simplified Chinese from the notebook.
notebooklm generate video "<title> video overview" --language zh_Hans --format explainer --style auto --no-wait --json
Wait ~5 min — check back and download the video via /get-notebooklm-video.
Step 3: Download Video Artifact
/skill get-notebooklm-video
Downloads video to /input/video.mp4 with title.json.
Step 4: Prepare for Upload
Moves video and title into the structure expected by upload scripts.
/skill notebooklm-prep-upload
Step 5: Publish to Platforms
Uploads to Bilibili, Douyin, Kuaishou, Weixin Video, or YouTube.
/skill pva-publisher
Quick Reference
| Skill | What it does |
|---|
/pva-check | Verify all prerequisites |
/setup-pva-notebooklm | This overview |
/notebooklm-research | Research a topic via NotebookLM |
/notebooklm-video | Research + generate video |
/get-notebooklm-video | Download video artifact |
/notebooklm-prep-upload | Prepare video for upload |
/pva-publisher | Cross-platform upload |
Note: Each step launches background work and resolves immediately. The user should wait the indicated time before proceeding to the next step.
