Sezzlee Local LLM MCP Server

Connect this server

This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.

claude mcp add llm -- npx -y @sezzlee/llm-mcp

{
  "mcpServers": {
    "llm": {
      "args": [
        "-y",
        "@sezzlee/llm-mcp"
      ],
      "command": "npx"
    }
  }
}

[mcp_servers.llm]
command = "npx"
args = ["-y", "@sezzlee/llm-mcp"]

{
  "mcpServers": {
    "llm": {
      "args": [
        "-y",
        "@sezzlee/llm-mcp"
      ],
      "command": "npx"
    }
  }
}

{
  "mcpServers": {
    "llm": {
      "args": [
        "-y",
        "@sezzlee/llm-mcp"
      ],
      "command": "npx"
    }
  }
}

This one needs environment variables set before it will start: SEZZLEE_LLM_MODEL (Ollama model to call), SEZZLEE_LLM_BASE_URL (Ollama base URL), SEZZLEE_LLM_NUM_CTX (Context window in tokens), SEZZLEE_LLM_TIMEOUT_MS (Per-call timeout in milliseconds), SEZZLEE_LLM_KEEP_ALIVE (How long Ollama keeps the model loaded), SEZZLEE_LLM_ROOT (Workspace folder paths resolve against), SEZZLEE_LLM_OUTPUT_DIR (Folder local_map writes its CSV into). The author declared them in the registry entry; get the values from the project itself.