One public endpoint gives an agent live MCP discovery and evidence before it connects to an unfamiliar server.
Universal endpoint
https://mcpqueen.com/mcp
Transport: Streamable HTTP · Authentication: none · Six read-only discovery/evidence tools plus one quarantined feedback tool.
Connect the universal URL as a private ChatGPT plugin/connector, or use it as a hosted MCP tool in the Responses API.
{
"type": "mcp",
"server_label": "mcpqueen",
"server_url": "https://mcpqueen.com/mcp"
}
Claude Code supports the remote HTTP endpoint directly. No local server or Docker image is required.
claude mcp add --transport http mcpqueen https://mcpqueen.com/mcp
Load the remote tools with MultiServerMCPClient. The example calls search_servers directly, so it needs no LLM key.
client = MultiServerMCPClient({
"mcpqueen": {
"transport": "http",
"url": "https://mcpqueen.com/mcp"
}
})
Use BasicMCPClient for direct calls, or McpToolSpec to give MCP Queen tools to a LlamaIndex agent.
client = BasicMCPClient(
"https://mcpqueen.com/mcp"
)
result = await client.call_tool(
"search_servers", {"query": "postgres"}
)
Call addMcpServer() from an Agent, then expose the read-only subset through this.mcp.getAITools().
async onStart() {
await this.addMcpServer(
"mcpqueen",
"https://mcpqueen.com/mcp"
);
}
The huggingface_hub.Agent can connect over HTTP and allowlist MCP Queen's read-only tools. Model-provider credentials remain separate.
{
"type": "http",
"config": {
"url": "https://mcpqueen.com/mcp"
}
}
Find → inspect evidence → decide → connect direct.
search_servers or search_tools for the task.get_server_grade and get_trust_receipt for candidates.Does MCP Queen require an API key?
No. The MCP Queen endpoint is public, uses Streamable HTTP, and requires no authentication. An agent framework may still require credentials for its own model provider.
Can an agent connect directly to a server it finds?
Yes. MCP Queen returns the server's published endpoint and evidence so the developer or agent can make a separate, explicit decision to connect directly.
Does an A operational grade prove that an MCP server is secure?
No. Operational grades measure protocol behavior, tool metadata, latency, and provenance observations. Security, data integrity, citations, claims, response benchmarks, and reviewed field use remain separate evidence dimensions; missing evidence is unaudited.
All runnable examples are maintained in the MCP Queen repository. The integration validator checks required artifacts, icon dimensions, submission structure, public URLs, live MCP tools, output schemas, and safety annotations without automating identity or final-review gates.