← Back to Skills Marketplace
4094
Downloads
1
Stars
58
Active Installs
1
Versions
Install in OpenClaw
/install web-access
Description
Universal web access skill: search, fetch, browser automation via CDP Proxy. Handles login-required sites, anti-scraping bypasses, and complex web interactions.
Usage Guidance
Install only if you intentionally want an agent to operate your real Chrome profile. Treat it like granting access to sites where you are already logged in. Avoid banking, admin consoles, private documents, sensitive accounts, and untrusted pages; require explicit approval before uploads, screenshots, purchases, account changes, or use of tokenized URLs; and stop the proxy when finished if you do not want a localhost browser-control API left running.
Capability Assessment
Purpose & Capability
The stated purpose openly includes web access, CDP browser automation, login-required sites, and anti-scraping handling, but the implemented proxy also exposes arbitrary page JavaScript, navigation, screenshots, real mouse clicks, and local-file upload into the user's everyday Chrome profile.
Instruction Scope
The instructions encourage use of authenticated browser state, preserving session-bearing URLs, bypass-oriented interaction tactics, and parallel sub-agents without clear per-site, per-action, or sensitive-account limits.
Install Mechanism
Installation/startup uses declared Node and curl dependencies and a helper script that starts a localhost CDP proxy; this is coherent with the skill, but users should know it creates a long-running local browser-control service.
Credentials
The proxy attaches to the user's normal Chrome session and inherits existing logins. It binds to 127.0.0.1, which limits network exposure, but the API has no authentication for local callers.
Persistence & Privilege
The skill recommends keeping the proxy running, allows screenshots to be saved to caller-supplied paths, accepts arbitrary local file paths for uploads, and instructs agents to write site-pattern knowledge files after browsing.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install web-access - After installation, invoke the skill by name or use
/web-access - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of the universal web-access skill.
- Provides search, fetch, and browser automation via CDP Proxy.
- Supports login-required sites, anti-scraping bypasses, and complex web interactions.
- Prioritizes user goals with adaptive, evidence-driven decision-making.
- Offers multiple web access modes: search, fetch, raw HTML, and full browser automation.
- Concurrent multi-agent workflows supported to handle independent subtasks efficiently.
- Site-specific operational knowledge maintained for reliable automation across different platforms.
Metadata
Frequently Asked Questions
What is Web Access Clawhub?
Universal web access skill: search, fetch, browser automation via CDP Proxy. Handles login-required sites, anti-scraping bypasses, and complex web interactions. It is an AI Agent Skill for Claude Code / OpenClaw, with 4094 downloads so far.
How do I install Web Access Clawhub?
Run "/install web-access" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Web Access Clawhub free?
Yes, Web Access Clawhub is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Web Access Clawhub support?
Web Access Clawhub is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Web Access Clawhub?
It is built and maintained by Eze (@eze-is); the current version is v1.0.0.
More Skills