A containerized web automation tool that utilizes AI vision to navigate complex sites and bypass graphical CAPTCHAs without external services.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-vision-scraper
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 agent-vision-scraper 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 Dockerized Vision Agent Scraper?
The Dockerized Vision Agent Scraper is a sophisticated tool designed for the Openclaw Skills ecosystem to handle human-like web operations. It operates within an isolated Docker container, featuring a built-in virtual screen (Xvfb) and Playwright stealth configurations to bypass modern anti-bot detection. Unlike traditional scrapers, this skill relies entirely on AI vision capabilities to read and solve graphical CAPTCHAs, eliminating the need for third-party decoding services.
By running in a dedicated environment, it ensures that even headless servers can execute tasks requiring a headed browser. This skill is particularly valuable for developers who need to automate interactions with highly secured or dynamic web interfaces while maintaining a low footprint and high reliability through the Openclaw Skills framework.
Dockerized Vision Agent Scraper Use Cases
- Automated data extraction from websites protected by graphical CAPTCHAs.
- Human-like interaction for automated account logins and form submissions.
- Scraping dynamic content from platforms with aggressive anti-scraping measures.
- Visual regression testing and UI verification in a containerized environment.
- Remote monitoring of automated browser sessions via VNC.
How Dockerized Vision Agent Scraper Works
- The user provides a target URL and a natural language instruction describing the desired task.
- The Openclaw Skills agent initiates a Docker container using a pre-configured Playwright image.
- A virtual screen (Xvfb) is started to support a graphical browser instance in a headless environment.
- Playwright Stealth is activated to mask the automation signature and avoid detection.
- The AI Vision model analyzes the browser viewport to identify elements, read text, and solve CAPTCHAs.
- The task is executed (e.g., clicking, typing, or scraping), and the resulting data or status is returned to the user.
Dockerized Vision Agent Scraper Setup
To use this within your Openclaw Skills library, you must first build the image and then execute it via Docker. Ensure your environment variables are configured in a .env file.
Build the image:
cd ~/.openclaw/skills/agent-vision-scraper
docker build -t agent-scraper-image .
Run the agent:
docker run --rm --env-file .env -p 5900:5900 agent-scraper-image node agent-scraper.js "URL" "Instruction"
Dockerized Vision Agent Scraper Data Schema & Taxonomy
The skill manages data through environmental configuration and runtime arguments as follows:
| Component | Description |
|---|---|
| Input Arguments | Accepts a target URL string and a task-specific prompt. |
| Environment Variables | Required .env file containing API keys for Vision LLMs. |
| Network Ports | Maps port 5900 for VNC remote desktop access to the browser session. |
| Execution Logs | Provides real-time console output of the scraping and vision reasoning steps. |
| Base Image | Utilizes mcr.microsoft.com/playwright:v1.58.2-jammy for consistent dependencies. |
Dockerized Vision Agent Scraper Advanced Features
- Real-time visual monitoring via VNC port 5900 connectivity.
- Advanced Playwright Stealth integration to bypass sophisticated bot detection systems.
- Autonomous graphical CAPTCHA solving using pure AI vision reasoning.
- Xvfb virtual display support for running GUI-based browsers on CLI-only servers.
- Seamless integration with existing Openclaw Skills workflows for multi-step data processing.
SKILL.md
Loading
Related Openclaw Skills
Microsoft Graph (Outlook & Calendar)
A comprehensive Microsoft Graph integration for managing Outlook email and calendar events through AI-driven automation.
Lingxi Real-Time Market Data Skill
A professional financial data integration providing real-time stock market metrics across A-shares, HK, and US markets for AI agents.
MiniMax Token Plan Tool
A lightweight JavaScript implementation for MiniMax API integration providing web search, vision-language analysis, and quota management.
hs (HelpScout CLI)
A comprehensive CLI tool to interact with HelpScout Inbox and Docs APIs for seamless support and knowledge management.
Engram MCP Server
A high-performance Model Context Protocol server that injects switchable expert memory and dynamic workflows into AI agents.
ES Search Skill
An AI-driven search utility that translates natural language queries into structured Elasticsearch API calls for comprehensive news and social media monitoring.
Featured*
