Ollama Handoff MCP Server

by michael-whitecapdataio.github.Michael-WhiteCapData/ollama-handoffv0.1.2

Offload summaries, drafts, extractions and first-pass code reviews from a cloud LLM agent to a local Ollama model via handoff tools with built-in prompts.

3Pythonstdioofficial registry

context tax

queued

security

queued

cold start

queued

freshness

Active16d ago

Install Ollama Handoff MCP server

claude mcp add ollama-handoff -- uvx ollama-handoff

{
  "mcpServers": {
    "ollama-handoff": {
      "command": "uvx",
      "args": [
        "ollama-handoff"
      ]
    }
  }
}

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project).

{
  "mcpServers": {
    "ollama-handoff": {
      "command": "uvx",
      "args": [
        "ollama-handoff"
      ]
    }
  }
}

Settings → Developer → Edit Config (claude_desktop_config.json), then restart.

{
  "servers": {
    "ollama-handoff": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "ollama-handoff"
      ]
    }
  }
}

Add to .vscode/mcp.json in your workspace.

{
  "mcpServers": {
    "ollama-handoff": {
      "command": "uvx",
      "args": [
        "ollama-handoff"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json.

Configuration

Variable	Required	Secret	Description
OLLAMA_URL	—	—	Base URL of the Ollama server.
OLLAMA_DEFAULT_MODEL	—	—	Default model used for handoffs.
OLLAMA_NUM_CTX	—	—	Context window in tokens.
OLLAMA_KEEP_ALIVE	—	—	How long to keep the model resident in VRAM.
OLLAMA_TIMEOUT_S	—	—	Per-request timeout in seconds.