Model Manager for VS Code
io.github.setheerwagen/mcp-modelmanager
Manage your local model machine over MCP: state, model pull/remove, service switch, LoRA training
client:VS Code
transport:stdio
runtime:pypi
Install Model Manager in VS Code
.vscode/mcp.json
{
"inputs": [
{
"type": "promptString",
"id": "mm-vm-host",
"description": "MM_VM_HOST",
"password": true
},
{
"type": "promptString",
"id": "mm-vm-user",
"description": "MM_VM_USER",
"password": true
},
{
"type": "promptString",
"id": "mm-container-root",
"description": "MM_CONTAINER_ROOT",
"password": true
},
{
"type": "promptString",
"id": "mm-max-tier",
"description": "MM_MAX_TIER",
"password": true
},
{
"type": "promptString",
"id": "mm-enabled-operations",
"description": "MM_ENABLED_OPERATIONS",
"password": true
},
{
"type": "promptString",
"id": "mm-vm2-host",
"description": "MM_VM2_HOST",
"password": true
}
],
"servers": {
"mcp-modelmanager": {
"type": "stdio",
"command": "uvx",
"args": [
"mcp-modelmanager"
],
"env": {
"MM_VM_HOST": "${input:mm-vm-host}",
"MM_VM_USER": "${input:mm-vm-user}",
"MM_CONTAINER_ROOT": "${input:mm-container-root}",
"MM_MAX_TIER": "${input:mm-max-tier}",
"MM_ENABLED_OPERATIONS": "${input:mm-enabled-operations}",
"MM_VM2_HOST": "${input:mm-vm2-host}"
}
}
}
}
VS Code prompts for each input the first time the server starts, so no secret is written to the file. VS Code docs