Read-only ops/health MCP server for a WSL2/systemd vLLM instance: liveness vs real-completion health tiers, GPU/VRAM, systemd state, live serve-flag introspection.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add vllm-ops -- uvx vllm-ops-mcp~/.codex/config.toml
[mcp_servers.vllm-ops]
command = "uvx"
args = ["vllm-ops-mcp"]
# codex mcp add vllm-ops -- uvx vllm-ops-mcp~/.cursor/mcp.json
{
"mcpServers": {
"vllm-ops": {
"command": "uvx",
"args": [
"vllm-ops-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"vllm-ops": {
"type": "stdio",
"command": "uvx",
"args": [
"vllm-ops-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"vllm-ops": {
"command": "uvx",
"args": [
"vllm-ops-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"vllm-ops": {
"command": "uvx",
"args": [
"vllm-ops-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"vllm-ops": {
"command": "uvx",
"args": [
"vllm-ops-mcp"
]
}
}
}