Guide for implementing WebMCP — the browser-native API that lets web pages expose structured tools to AI agents. Covers both the W3C standard (document.modelContext API, imperative and declarative approaches) and the original WebSocket-bridge implementation. Use this skill whenever the user asks about WebMCP, wants to add AI agent tools to a website, needs to expose web page functionality to browser agents, mentions 'document.modelContext', 'registerTool' in browser context, wants to create declarative tools from HTML forms, or is building any kind of browser-side MCP integration. Also use when the user mentions 'agentic web', 'tools for AI agents in the browser', or asks how to make a website work with AI assistants like Gemini, Claude, or ChatGPT browser agents.
$npx -y skills add biggora/claude-plugins-registry --skill web-mcpInstalls into the current project.
Run `npx skills use "https://github.com/biggora/claude-plugins-registry" --skill "biggora/claude-plugins-registry/web-mcp"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/biggora/claude-plugins-registry" that are relevant to the current task. Run `npx skills add "https://github.com/biggora/claude-plugins-registry"` and select the relevant skills, then follow their instructions.