Lightweight MCP client that uses a local Ollama LLM to query multiple MCP servers defined in config.json
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add ollama-nagharjun17 -- uv run client.py~/.codex/config.toml
[mcp_servers.ollama-nagharjun17]
command = "uv"
args = ["run", "client.py"]
# codex mcp add ollama-nagharjun17 -- uv run client.py~/.cursor/mcp.json
{
"mcpServers": {
"ollama-nagharjun17": {
"command": "uv",
"args": [
"run",
"client.py"
]
}
}
}.vscode/mcp.json
{
"servers": {
"ollama-nagharjun17": {
"type": "stdio",
"command": "uv",
"args": [
"run",
"client.py"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"ollama-nagharjun17": {
"command": "uv",
"args": [
"run",
"client.py"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"ollama-nagharjun17": {
"command": "uv",
"args": [
"run",
"client.py"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"ollama-nagharjun17": {
"command": "uv",
"args": [
"run",
"client.py"
]
}
}
}