MCP.so
Sign In
Servers
G

Gcore Cloud Mcp Server

@G-Core

Gcore MCP Server makes your AI assistant:

Overview

What is Gcore Cloud Mcp Server?

Gcore Cloud Mcp Server is an MCP server that exposes Gcore’s cloud API tools to LLM assistants, allowing them to manage cloud resources through predefined tool sets. It is intended for DevOps and cloud automation workflows.

How to use Gcore Cloud Mcp Server?

Install with uv tool install "gcore-mcp-server@git+https://github.com/G-Core/gcore-mcp-server.git" and run gcore-mcp-server. Configure resource categories via the GCORE_TOOLS environment variable (e.g., management,instances or wildcard patterns like cloud.instances.*). Point your LLM client at the HTTP endpoint or use stdio mode to start issuing MCP requests.

Key features of Gcore Cloud Mcp Server

  • Nine predefined toolsets covering cloud operations.
  • Support for HTTP and stdio transport modes.
  • Configurable tool selection via GCORE_TOOLS environment variable.
  • Wildcard patterns for fine-grained API access control.
  • Optional integration with MCP Inspector for debugging.

Use cases of Gcore Cloud Mcp Server

  • Spinning up GPU clusters for HPC and rendering workloads.
  • Auditing cloud resource costs and usage.
  • Tearing down test environments automatically.
  • Managing full VM lifecycle (create, stop, reboot, resize, snapshot).
  • Automating account, project, and permission management.

FAQ from Gcore Cloud Mcp Server

What toolsets does Gcore Cloud Mcp Server provide?

It provides nine predefined toolsets: management, instances, volumes, networks, baremetal, gpu_baremetal, cleanup, list, and more.

How do I control which tools the server exposes?

Set the GCORE_TOOLS environment variable with a comma-separated list of toolset names or wildcard patterns (e.g., management,instances or cloud.instances.*).

What transport modes are supported?

HTTP and stdio modes are supported. HTTP defaults to management,instances, and stdio defaults to management.

How do I install and run Gcore Cloud Mcp Server?

Install using uv tool install "gcore-mcp-server@git+https://github.com/G-Core/gcore-mcp-server.git" and then run gcore-mcp-server.

Can I debug the server’s interactions?

Yes, you can use MCP Inspector (npx @modelcontextprotocol/inspector) to explore tools, test calls, and monitor LLM-server traffic in real time.

Tags

More from Other