WebMCP Generator:
Make Your Site Accessible to MCP-Compatible AI Agents
Generate and deploy WebMCP configuration that makes your website accessible and actionable for MCP-compatible AI agents - including Claude, Cursor, and custom MCP clients. Turn any website into an agent-compatible tool endpoint.
WebMCP: The Protocol That Makes Websites Agent-Compatible
The Model Context Protocol (MCP) was designed for tool integrations in AI development environments. WebMCP extends this to the open web - allowing any website to declare its capabilities as MCP tools, making it discoverable and usable by any MCP-compatible AI agent without custom integration work.
Citerank's WebMCP Generator analyzes your site, identifies the actions and data endpoints you expose, and generates a spec-compliant WebMCP manifest. Deploy it at /.well-known/webmcp.json and your site becomes immediately actionable for Claude, Cursor, and any other MCP-capable agent.
How the WebMCP Generator Works
-
Enter Your Domain
The generator crawls your site to discover available endpoints, APIs, forms, and actions - building a complete picture of what your site offers to agents.
-
Tool and Endpoint Discovery
The generator identifies: search endpoints, form submission endpoints, data feeds, API endpoints (if documented), and potentialAction schema declarations - all candidates for WebMCP tool definitions.
-
Generate and Deploy the Manifest
Outputs a complete WebMCP manifest with tool definitions for every discovered endpoint. Includes one-click deployment instructions for common platforms - WordPress, Webflow, Next.js, and static site hosts.
What the WebMCP Generator Produces
-
Spec-Compliant WebMCP Manifest
Generates a well-formed WebMCP manifest at the correct path (/.well-known/webmcp.json) with accurate tool definitions for your site's capabilities.
-
Tool Definition Generation
Converts your site's actions and endpoints into MCP tool definitions - with name, description, input schema, and endpoint declarations that AI agents can use directly.
-
potentialAction Integration
Reads existing potentialAction schema on your pages and converts it into corresponding WebMCP tool definitions - ensuring consistency between your schema layer and your agent protocol layer.
-
Platform-Specific Deployment Guides
Provides step-by-step deployment instructions for WordPress, Webflow, Next.js, Vercel, and Apache/Nginx - so you can deploy the manifest regardless of your stack.
-
Agent Compatibility Check
After deployment, verifies that the manifest is correctly accessible and parseable by MCP-compatible clients - confirming your configuration is agent-ready.
-
Update Detection
Monitors your site for new endpoints or actions and alerts you when the WebMCP manifest should be updated to reflect new capabilities.
Who Uses the WebMCP Generator
- Developers making their web applications accessible to MCP-compatible AI agents
- SaaS companies exposing their product features to the agentic AI ecosystem
- E-commerce platforms enabling AI agents to search products and initiate purchases
- Service businesses allowing AI agents to book, quote, and contact on their behalf
- Content publishers exposing their search and archive endpoints to AI research agents
- Agencies implementing WebMCP as part of agentic readiness services
Frequently Asked Questions
What is WebMCP?
WebMCP is an extension of the Model Context Protocol (MCP) that enables any website to declare its capabilities as MCP tools via a manifest file at /.well-known/webmcp.json. AI agents using MCP-compatible clients can discover and call these tools automatically, making your website programmatically actionable without custom integration.
Which AI agents support WebMCP?
Claude (via Anthropic's MCP implementation), Cursor, and any AI agent built on an MCP-compatible client supports WebMCP. The ecosystem is expanding rapidly as MCP becomes the standard protocol for AI-web interaction.
Does my site need an API for WebMCP to work?
No. WebMCP can wrap form submissions, search endpoints, and other standard web interactions - not just REST or GraphQL APIs. The generator identifies whatever actionable endpoints your site exposes, regardless of their technical implementation.
Is WebMCP the same as llms.txt?
No. llms.txt provides context about your site's content for AI systems. WebMCP provides tool definitions that AI agents can call to take actions on your site. Both are complementary agentic protocols - llms.txt for understanding, WebMCP for doing.
What is the /.well-known/ path?
The /.well-known/ directory is a standardized web convention for placing machine-readable configuration files at predictable locations. Placing webmcp.json there allows AI agents to discover your WebMCP manifest without needing to be told the exact URL.
How do I verify my WebMCP configuration is working?
Citerank's Agent Compatibility Check fetches your manifest, validates its structure, and attempts to simulate tool calls against your declared endpoints - confirming that an AI agent could use your tools correctly.