Overview

Self Healing Infrastructure MCP provides AI-powered tools via the Model Context Protocol (MCP).

Tools

Tool	Description
node_health_check	Check health of infrastructure nodes including CPU, memory, disk, and GPU.
cluster_health_check	Check health of all 9 nodes in the cluster.
restart_service	Restart a failed service with automatic rollback on repeated failures.
cost_report	Generate infrastructure cost report with breakdown by service, region, and resou
auto_remediate	Automatically remediate common infrastructure issues using predefined playbooks.
gpu_orchestration	Orchestrate GPU allocation across the 9-node cluster.
failover_decision	Determine failover action for a failed node in the cluster.

Installation

pip install meok-self-healing-infrastructure-mcp

Usage with Claude Desktop

Add to your Claude Desktop MCP config (claude_desktop_config.json):

{
  "mcpServers": {
    "self-healing-infrastructure-mcp": {
      "command": "python",
      "args": ["-m", "meok_self_healing_infrastructure_mcp.server"]
    }
  }
}

Usage with FastMCP

from mcp.server.fastmcp import FastMCP

# This server exposes 7 tool(s) via MCP
# See server.py for full implementation

License

MIT © MEOK AI Labs