Runs supported ComfyUI image workflows on a local GPU with SHA-256 model identity checks, explicit license approvals and retained run evidence.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add local-gpu-imagegen -- uvx local-gpu-imagegen~/.codex/config.toml
[mcp_servers.local-gpu-imagegen]
command = "uvx"
args = ["local-gpu-imagegen"]
# codex mcp add local-gpu-imagegen -- uvx local-gpu-imagegen~/.cursor/mcp.json
{
"mcpServers": {
"local-gpu-imagegen": {
"command": "uvx",
"args": [
"local-gpu-imagegen"
]
}
}
}.vscode/mcp.json
{
"servers": {
"local-gpu-imagegen": {
"type": "stdio",
"command": "uvx",
"args": [
"local-gpu-imagegen"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"local-gpu-imagegen": {
"command": "uvx",
"args": [
"local-gpu-imagegen"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"local-gpu-imagegen": {
"command": "uvx",
"args": [
"local-gpu-imagegen"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"local-gpu-imagegen": {
"command": "uvx",
"args": [
"local-gpu-imagegen"
]
}
}
}