Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP-MCSTATUSCheck the status and current player count for mc.hypixel.net"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP-MCSTATUS
MCP server (Python) with tools for https://mcstatus.xyz/api.
Implemented MCP Tools
get_minecraft_status- Minecraft server status (Java/Bedrock), endpoint/api/statusget_java_status- Java status shortcut, endpoint/api/statusget_bedrock_status- Bedrock status shortcut, endpoint/api/statusget_srv_records- SRV records, endpoint/api/srvresolve_dns- DNS resolution and provider info, endpoint/api/dnsrdns- reverse DNS (PTR) lookup for IPgeoip_maxmind- GeoIP lookup using local MaxMind GeoLite2 databaseget_ip_provider_info- provider/operator info for IP viabgp.toolswhois + ASN databaseis_ip_anycast- check if player IP is Anycast by curated known-node listget_bgp_info- BGP/ASN details for an IP, endpoint/api/bgp
Architecture
mcstatus_mcp/client.py- typed API client (MCStatusApiClient)mcstatus_mcp/tools.py- abstractBaseMCStatusTool+ one class per MCP toolmcstatus_mcp/server.py- MCP app bootstrap and tool registration
Each tool is implemented as a class that inherits from BaseMCStatusTool.
All tools depend on a shared MCStatusApiClient instance.
Install
Run With Docker Compose
Stop:
Default MCP endpoint from compose:
http://localhost:8000/mcp
Run MCP Server
Default transport is stdio:
Optional transport override:
Allowed MCP_TRANSPORT values: stdio, sse, streamable-http.
Local Stdio Config Example
Environment Variables
MCP_TRANSPORT- MCP transport (default:stdio)MCSTATUS_API_BASE_URL- API base URL (default:https://mcstatus.xyz/api)MCSTATUS_TIMEOUT_MS- default timeout in milliseconds for tools (default:4000)MCP_HOST- host for HTTP transports (sseandstreamable-http, default:127.0.0.1)MCP_PORT- port for HTTP transports (default:8000)MCP_STREAMABLE_HTTP_PATH- streamable HTTP path (default:/mcp)MCP_SSE_PATH- SSE path (default:/sse)MAXMIND_LICENSE_KEY- MaxMind license key for GeoLite2 download (required for auto-download if DB is missing/outdated)MAXMIND_DB_PATH- local path to.mmdbfile (default:data/GeoLite2-City.mmdb)MAXMIND_EDITION_ID- MaxMind edition ID (default:GeoLite2-City)MAXMIND_REFRESH_HOURS- database refresh interval in hours;0disables periodic refresh (default:24)BGPTOOLS_USER_AGENT- descriptive user-agent with contact for downloadinghttps://bgp.tools/asns.csv(recommended)BGPTOOLS_ASN_DB_URL- ASN CSV source URL (default:https://bgp.tools/asns.csv)BGPTOOLS_ASN_DB_PATH- local path to ASN CSV cache (default:data/bgp_tools_asns.csv)BGPTOOLS_ASN_REFRESH_HOURS- ASN CSV refresh interval in hours;0disables periodic refresh (default:24)BGPTOOLS_WHOIS_HOST- bgp.tools whois host (default:bgp.tools)BGPTOOLS_WHOIS_PORT- bgp.tools whois port (default:43)
Use With OpenAI
Important: OpenAI MCP integration uses remote MCP servers.
Local stdio servers are good for local dev/testing, but for OpenAI binding you should expose a public HTTPS endpoint.
1) Run as streamable HTTP for deployment
After deployment your MCP URL will look like:
https://your-domain.example/mcp
2) Bind MCP server in OpenAI Responses API
3) Bind in ChatGPT (Connectors)
Open ChatGPT connector settings.
Add a custom connector using your remote MCP URL.
Select tools and permissions.
Reference docs:
https://platform.openai.com/docs/guides/tools-remote-mcp
https://platform.openai.com/docs/guides/mcp
https://help.openai.com/en/articles/11487775-connectors-in-chatgpt/