A specialized tool for AI agents to fetch and process transcripts from Bilibili videos for summarization and analysis.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bilibili-watcher
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install bilibili-watcher using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Bilibili Watcher?
Bilibili Watcher is an essential utility for developers and AI agents designed to bridge the gap between Bilibili video content and text-based analysis. By leveraging advanced extraction techniques, this tool allows users to pull accurate transcripts from B 站 videos, facilitating deeper content understanding without needing to watch the video manually. As part of the Openclaw Skills ecosystem, it empowers AI assistants to interact with one of the world's largest video platforms through direct data retrieval.
This skill is particularly useful for those who need to process high volumes of educational or informative video content quickly. By transforming audio-visual data into searchable text, it allows for more efficient knowledge management and better integration into automated coding or research workflows.
Bilibili Watcher Use Cases
- Summarizing lengthy educational videos or technical tutorials on Bilibili.
- Extracting specific technical data, code snippets, or instructions from video transcripts.
- Enabling AI agents to answer complex questions based on B 站 video content.
- Facilitating content research by making video information searchable through text.
How Bilibili Watcher Works
- The AI agent receives a request involving a Bilibili video URL or a shortened b23.tv link.
- The skill invokes a Python script that utilizes the yt-dlp library to interface with Bilibili's metadata API.
- The system checks the video metadata for the presence of manual closed captions (CC) or auto-generated subtitles.
- If subtitles are found, they are extracted, cleaned, and returned as a plain text transcript for the agent to analyze.
Bilibili Watcher Setup
To get started with this skill, you must ensure the core dependency is installed on your system. This skill requires yt-dlp to be available in your PATH.
Install the dependency via pip:
pip install yt-dlp
You can then test the transcript retrieval using the following command structure:
python3 scripts/get_transcript.py "https://www.bilibili.com/video/BV1xx411c7mD"
Bilibili Watcher Data Schema & Taxonomy
The skill manages video identifiers and outputs text data. Below is the structure of how information is processed:
| Data Point | Description |
|---|---|
| Input URL | Supports BV, AV, and b23.tv shortened URL formats. |
| Output Type | Raw text transcript or error message if no subtitles exist. |
| Dependencies | Requires yt-dlp for backend video communication. |
| Triggers | Configured to respond to terms like watch bilibili or b 站总结. |
Bilibili Watcher Advanced Features
- Comprehensive URL support covering modern BV IDs, legacy AV IDs, and mobile-friendly short links.
- Seamless integration with the Openclaw Skills framework for multi-agent research pipelines.
- Capability to handle both auto-generated and manual captions for maximum coverage.
- Robust error handling that provides specific feedback when a video lacks subtitle tracks.
SKILL.md
Loading
Related Openclaw Skills
Doc Sync
Doc Sync is an automated documentation assistant that ensures code comments, README files, and local knowledge bases remain perfectly synchronized with your codebase.
Insurance Policy Reader
An AI-powered document analyzer that translates complex insurance policies into plain language summaries highlighting coverage, costs, and exclusions.
docx-tables
A specialized technical workflow for generating programmatically perfect Word tables that maintain consistency across Microsoft Word and Google Docs.
Idea-to-Post Expansion
An AI-driven content expansion engine that uses progressive questioning and multi-stage web search to turn raw ideas into high-impact social media posts.
FAQ Generator From Reviews
Transform e-commerce customer feedback and objections into authentic, data-driven FAQ content for product pages and support desks.
YoudaoNote CLI
A robust command-line interface for YoudaoNote that enables terminal-based note management, web clipping, and workflow automation.
Featured*
