webmcp
This skill should be used when browsing or automating web pages that expose tools via the WebMCP API (window.navigator.modelContext). It teaches agents how to discover, inspect, and invoke WebMCP tools on websites instead of relying on DOM scraping or UI actuation.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/brunobuddy/webmcpWhat This Skill Does
The webmcp skill enables OpenClaw agents to interact directly with websites that implement the WebMCP standard via window.navigator.modelContext. Unlike traditional browser automation which relies on brittle DOM selectors, coordinate-based clicking, or visual analysis, WebMCP provides a structured API for agents to discover and invoke functionality exposed by web developers. By utilizing this skill, agents act as native users of the application, invoking functions that can modify state, submit forms, or trigger background tasks with the reliability of direct function execution. It essentially turns every WebMCP-enabled page into a programmable interface for your AI agent.
Installation
To integrate this capability into your environment, use the OpenClaw command-line interface or agent console to install the skill package from the official registry:
clawhub install openclaw/skills/skills/brunobuddy/webmcp
Once installed, the agent will automatically attempt to detect the modelContext API when navigating to any web URL. No complex configuration files are required, but ensure the agent has the necessary permissions to execute browser-based tasks.
Use Cases
- Complex Form Filling: Instead of guessing where input fields are, use a registered
submit-ordertool to perform operations reliably. - SaaS Workflow Automation: Execute high-level commands on enterprise dashboards that expose internal APIs to the AI.
- Deep SPA Interaction: Interact with stateful single-page applications where the UI might not reflect underlying data structures, bypassing visual UI synchronization issues.
- Data Exporting: Trigger built-in data processing or export tools provided by the website owner to extract information in structured formats.
Example Prompts
- "Check if the current page supports WebMCP and list all available tools for managing my cloud storage bucket."
- "Use the 'create-document' tool from the page's modelContext to draft a new project proposal with the title 'Q4 Roadmap'."
- "The current site has a 'sync-settings' tool; please invoke it with the parameters provided in the documentation to ensure my account profile is up to date."
Tips & Limitations
- Discovery First: Always perform a check for
window.navigator.modelContextbefore attempting to use tools. If not found, revert to standard vision-based navigation. - SPA Sensitivity: Single-page applications can register or unregister tools dynamically based on user navigation. Re-scan for tools whenever the page route changes.
- Security Context: WebMCP tools execute on the page's main thread. Only use this skill on trusted domains, as malicious sites could expose dangerous tools via this API. Always review the tool descriptions before invoking them.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-brunobuddy-webmcp": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: code-execution, external-api
Related Skills
Startup Info
Skill by brunobuddy
manifest
Model Router for OpenClaw. Save up to 70% by routing requests to the right model. Choose LLM fallback to avoid API rate limits, set thresholds and reduce token consumption.
hide-openclaw-banner
Suppress the OpenClaw CLI banner message. Use when the user says "hide openclaw banner", "disable openclaw banner", "suppress openclaw banner", "remove openclaw banner", "openclaw banner is annoying", "turn off the lobster", or wants to stop seeing the OpenClaw startup tagline on every command invocation.