Web Pilot for Openclaw

An API-free web searching and interactive browsing toolset powered by Playwright for AI agents.

liranudi
v1.0.0
Feb 18, 2026
1
10.2k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install web-pilot

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install web-pilot 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 Web Pilot?

Web Pilot is a comprehensive suite of scripts designed to empower AI agents with the ability to navigate the live web, search multiple engines, and interact with complex page elements without requiring expensive API keys. As a robust entry in the Openclaw Skills collection, it provides a bridge between static LLM knowledge and real-time internet data by leveraging Playwright and Chromium for authentic browser emulation.

The tool is built for versatility, offering everything from simple one-shot page reading to complex, persistent browser sessions. By using Web Pilot within the Openclaw Skills ecosystem, developers can automate tasks like clicking buttons, switching between multiple browser tabs, and dismissing intrusive cookie banners, ensuring that the AI agent receives clean, structured data in JSON, Markdown, or plain text formats.

Web Pilot Use Cases

  • Conducting multi-engine web searches to gather diverse data points for research.
  • Extracting clean, readable text from complex URLs for LLM summarization.
  • Automating interactions with web applications that require multi-step navigation or clicking.
  • Downloading and automatically extracting text from online PDF documents.
  • Monitoring web content changes via automated screenshots and text searches within pages.

How Web Pilot Works

  1. The agent initiates a search across DuckDuckGo, Brave, or Google to identify relevant URLs.
  2. A Playwright-powered Chromium instance is launched to navigate to the target web page.
  3. The skill automatically identifies and dismisses cookie consent banners to access the primary content.
  4. Page content is parsed and converted into the requested format (JSON, Markdown, or Text) for the agent to process.
  5. If a persistent session is required, the browser remains open to allow for subsequent actions like tab switching, clicking, or taking screenshots.

Web Pilot Setup

To integrate this skill into your environment, install the core dependencies and the Chromium browser engine:

pip install requests beautifulsoup4 playwright
playwright install chromium

For enhanced functionality like PDF text extraction, you can optionally install additional libraries:

pip install pdfplumber

Web Pilot Data Schema & Taxonomy

Web Pilot ensures high data interoperability within the Openclaw Skills framework by returning strictly structured outputs.

Feature Output Type Primary Metadata Fields
Web Search JSON List title, url, snippet
Page Reader JSON/Markdown content, url, status
File Downloader JSON filename, size_bytes, extracted_text
Browser Session JSON active_tab_index, page_title, search_results

Web Pilot Advanced Features

  • Persistent browser sessions that maintain state across multiple navigation commands.
  • Multi-tab management allowing agents to open, switch, and close independent pages.
  • Automated cookie banner dismissal for seamless data extraction from modern websites.
  • Integrated PDF processing that automatically converts downloads into searchable text.
  • Visual verification through full-page or viewport screenshots during automated workflows.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*