ACCESS-CI MCP Servers

ACCESS-CI MCP ServersConnect your AI assistant to ACCESS-CI resources, communities, and data.

Public Model Context Protocol servers — query compute resources, check system status, search software, explore allocations, and more through natural conversation.

Get Started
View on GitHub

What is MCP? ​

The Model Context Protocol (MCP) is an open standard that lets AI assistants securely interact with external tools, systems, and data sources. It provides a universal interface for AI to read data, call functions, and reason over real-world context. MCP is supported by Claude and a growing number of AI tools.

What You Can Do ​

With ACCESS-CI MCP servers connected, you can ask your AI questions like:

  • "What GPU resources are available on ACCESS-CI?"
  • "Are there any current outages or upcoming maintenance on Delta?"
  • "Show me upcoming workshops on parallel computing"
  • "What machine learning software is available on Anvil?"
  • "What NSF-funded projects are doing work in quantum computing?"

Your AI queries the relevant ACCESS services and gives you a synthesized answer — no need to navigate multiple websites or APIs.

Available Servers ​

ACCESS-CI MCP servers provide structured access to:

  • Compute Resources — Hardware specifications, GPU availability, system capabilities
  • System Status — Outages, maintenance schedules, and incidents
  • Software Discovery — Software packages available across ACCESS resources
  • XDMoD — Usage statistics, visualizations, and resource utilization
  • Allocations — Research projects, allocation trends, and collaboration discovery
  • NSF Awards — NSF funding data and cross-referencing
  • Documentation — ACCESS-CI documentation retrieval (how-tos, policies, guides)
  • Events — Workshops, webinars, and training sessions
  • Announcements — Community news and updates
  • Affinity Groups — Communities, events, and knowledge base resources

A tenth server, XDMoD Data Analytics, provides per-user job and usage data via Claude Code (it requires API headers that aren't supported by Claude.ai connectors). Details →

Quick Start ​

Most tools work without authentication — adding a server takes about a minute.

Claude Desktop / claude.ai ​

  1. Open Customize > Connectors > Add custom connector
  2. Paste a server URL — start with https://mcp.access-ci.org/compute-resources/sse
  3. Authorize when prompted

You can add as many or as few servers as you need. See the Getting Started guide for all server URLs.

Claude Code ​

See the Getting Started guide for CLI commands and configuration.

Other MCP Clients ​

The Model Context Protocol is supported by a growing number of AI tools. ACCESS MCP servers should work with any client that supports remote MCP servers — see the Other MCP Clients section for details. We'd love to hear about your experience: open a support ticket so we can expand our documentation.

Self-Hosting ​

Want to run the servers yourself? See the setup guide for Docker deployment instructions.

Supported by ACCESS-CI ​

These MCP servers are built to work with official ACCESS-CI APIs and are designed to help researchers more effectively discover and utilize cyberinfrastructure resources.