← Back to Skills Marketplace
118
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install web-automation
Description
Automate browsing tasks with advanced navigation, interaction, data scraping, form filling, and high-res screenshot capture using a headless browser.
Usage Guidance
This skill's description and runtime doc promise a headless-browser automation engine but provide no implementation, dependency list, or secret-handling guidance. Before installing or using it, ask the publisher: (1) what runtime/binary provides the headless browser (Chromium/Playwright/Puppeteer/etc.), (2) how the agent obtains and runs that browser (preinstalled runtime, remote service, or package install), and (3) how credentials should be provided and protected during 'login' tasks (do not paste primary account passwords into chat). If you must test it, use a non-sensitive test account and limit scope (no bank, work, or personal accounts). Prefer skills that include explicit install steps, declared dependencies, and secure handling instructions for secrets.
Capability Analysis
Type: OpenClaw Skill
Name: web-automation
Version: 1.0.0
The skill bundle contains standard metadata and documentation for a web automation tool. The instructions in SKILL.md describe typical browser capabilities such as navigation, screenshotting, and form interaction, which are consistent with the stated purpose and show no signs of malicious intent or prompt injection.
Capability Tags
Capability Assessment
Purpose & Capability
Name and description claim a full headless-browser automation tool (navigation, JS interaction, screenshots, form logins). However the skill bundle contains no code, no install instructions, and declares no required binaries (Chrome, Chromium, Playwright, Puppeteer, Selenium, etc.). That mismatch means the skill either relies on an undocumented runtime component or is incomplete; either is unexplained and inconsistent with the claimed capabilities.
Instruction Scope
The SKILL.md is high-level and open-ended: it tells agents to use the skill for browsing and to 'login to a portal' but provides no concrete runtime commands, no API endpoints, and no guidance for securely obtaining or storing credentials. This grants the agent broad discretion (which pages to visit, what credentials to use) without safety or scope limits.
Install Mechanism
There is no install spec and no files beyond SKILL.md, so the skill does not write code to disk (low install risk). However, the doc explicitly references a headless browser engine 'optimized for OpenCloudOS' — claiming a runtime dependency that is not declared or documented is an implementation inconsistency to be clarified.
Credentials
The skill requests no environment variables or credentials (proportionate on its face). But because the intended use includes automated logins and form-filling, the lack of any guidance about how credentials should be supplied, validated, or protected is a concern: sensitive secrets may be handled ad hoc in conversation or injected without clear rules.
Persistence & Privilege
always is false and there is no claim of modifying system or other skills' configs. Autonomous invocation remains allowed (platform default) but this alone is not a new risk here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install web-automation - After installation, invoke the skill by name or use
/web-automation - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of comprehensive web automation and browser control features for OpenClaw agents.
- Navigate to URLs, handle redirects, and manage session cookies.
- Capture high-resolution screenshots of full pages or specific elements.
- Interact with JavaScript-rendered elements using clicks and hovers.
- Extract text, attributes, and structured data from web pages.
- Automate multi-step forms, dropdown selections, and submissions.
- Built-in retry logic and support for custom user-agent strings.
Metadata
Frequently Asked Questions
What is Web Automation?
Automate browsing tasks with advanced navigation, interaction, data scraping, form filling, and high-res screenshot capture using a headless browser. It is an AI Agent Skill for Claude Code / OpenClaw, with 118 downloads so far.
How do I install Web Automation?
Run "/install web-automation" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Web Automation free?
Yes, Web Automation is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Web Automation support?
Web Automation is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Web Automation?
It is built and maintained by rudigta (@rudigta); the current version is v1.0.0.
More Skills