⬡ What is MCP?
The Model Context Protocol is an open standard that lets AI assistants call external tools mid-conversation. An MCP server exposes a set of named tools — each with a description and input schema — that any compatible AI can discover and use automatically.
This registry lists public MCP servers so anyone can find them. No central authority, no approval queue. If your server responds to a tools/list request, it gets verified automatically on registration.
POST /mcp-registry/submit
📡 Registered servers …
+ Register your server
Paste your MCP endpoint URL. We'll ping it immediately with a tools/list request to verify it's live and pull your tool names automatically. Even if the ping fails, you're listed — just marked unverified until the endpoint responds.
⚡ Register via API
Any AI or script can register an MCP server programmatically — no browser required:
POST https://betterthanhtml.com/mcp-registry/submit
Content-Type: application/json
{
"name": "My MCP Server",
"url": "https://example.com/mcp",
"description": "What it does",
"author": "Your name or @handle"
}
Response:
{
"ok": true,
"id": "example-com-mcp",
"verified": true,
"tools": ["tool_one", "tool_two"],
"message": "Registered and verified. 2 tools found: tool_one, tool_two."
}
The registry JSON is publicly readable at GET /mcp-registry — suitable for inclusion in llms.txt, system prompts, or any AI context that needs to know what MCP servers are available.