A versatile automation capability for Openclaw Skills that extracts trending Douyin videos, hot list metrics, and creative copy using browser replication or scripting.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install douyin-scraper-2
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 douyin-scraper-2 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 Douyin Scraper?
The Douyin Scraper is a technical enhancement tailored for the Openclaw Skills ecosystem, enabling AI agents to seamlessly extract valuable media assets from Douyin (TikTok China). By translating natural language inputs into structured automation commands, it retrieves video metrics, trending lists, and video descriptions directly from target pages.
To ensure reliable execution, this skill uses dual access modes: live browser automation via OpenClaw's native engine, and direct fallback scripting using Playwright. It is optimized to support authenticated user profiles, which drastically decreases platform rate limiting and verification blocks, delivering structured JSON payloads ready for analytics or curation.
Douyin Scraper Use Cases
- Market Research & Curation: Auto-extract top viral copy and popular tags around specific industries like food, gaming, or lifestyle.
- Trend Intelligence: Programmatically monitor the real-time Douyin hot list to spot explosive topics before they saturate the market.
- Performance Benchmarking: Aggregate video statistics (views, likes, and creator handles) to evaluate competitors' reach.
- Dynamic Media Workflows: Combine content discovery with automation pipelines built using Openclaw Skills.
How Douyin Scraper Works
- Natural Language Decoding: The AI parses conversational phrases (e.g., "check Douyin hot list") to determine the target action (
search,hot, orvideo), extraction keywords, and item limits. - Browser Environment Orchestration: The skill launches the OpenClaw browser tool with
profile="user"to utilize logged-in browser sessions, lowering detection risk. - Extraction & Dom Snapshotting: The agent reads page snapshots, analyzes structural components, and parses raw video cards into clean data nodes.
- Fallback & Recovery: If captchas trigger or headless execution is blocked, the script degrades gracefully to simulated mock datasets to prevent application crashes.
Douyin Scraper Setup
Configure your scraping backend dependencies within your project workspace to leverage this capability alongside your other Openclaw Skills.
# Install the browser dependencies (Python/Playwright)
pip install playwright
playwright install
# Or install Node.js dependencies if using the JavaScript runner
npm install playwright
Run verification tasks directly from the command line:
# Execute keyword search through the Python runner
python3 scripts/scraper.py search --keyword "海鲜" --limit 10
# Retrieve current trending topics using mock safeguards
python3 scripts/scraper.py hot --limit 20 --mock
Douyin Scraper Data Schema & Taxonomy
All extracted video datasets are converted into a standardized JSON model. Below is the schematic overview of the metadata taxonomy:
| Property | Data Type | Description |
|---|---|---|
title |
String | The main headline, caption, or video description text |
author |
String | Creator's unique username or display name |
play_count |
Integer | Total registered view count of the video |
like_count |
Integer | Total likes received by the video |
url |
String | Direct, canonical URL address pointing to the video page |
tags |
Array | Curated hashtag strings extracted from the video description |
Douyin Scraper Advanced Features
- Authenticated Session Persistence: Integrates with local user profiles (
profile="user") to bypass aggressive initial platform logins and secure human-like browse states. - Anti-Failure Mock Fallback: Includes safety-first mock loops in script engines that output reference data instead of hard-throwing exceptions when captchas block headless browsers.
- Natural Language Parsing Hooks: Allows non-technical operators to trigger deep-dive scrapes using simple conversational speech commands.
- Dual Language Support: Ready-to-use boilerplate controllers in both Python and Node.js to fit whatever tech stack you build on.
SKILL.md
Loading
Related Openclaw Skills
Art Critique
An AI-powered creative companion that provides profound, non-corrective aesthetic critique for visual, textual, audio, and performance art, backed by art theory and structured dialogue tracking.
Exam Practice Site Generator
A data-driven, single-page exam practice site generator that decouples interactive frontends from JSON question banks, enabling effortless adaptation to any standardized test.
Property Graph Schema Designer
An AI agent utility that translates domain models, requirements, and entity descriptions into structured Neo4j-style property graph schemas.
Craw Figma
A bidirectional integration enabling AI agents to query layouts via the cloud REST API and write to, edit, or audit layers via a local WebSocket-based plugin.
Code Mirror
A powerful full-stack code generation and synchronization skill that mirrors APIs, data schemas, validation rules, and client logic between frontend and backend stacks.
Huo15 Odoo Integration
A zero-dependency Python toolset facilitating task tracking, timesheet aggregation, CRM administration, and calendar reminders on Odoo through XML-RPC.
Featured*
