MCP Server
MCP
io.github.Michael-WhiteCapData/ollama-handoff
Offload cheap work from your AI agent to a local Ollama model, at zero cloud cost.
Install
uvx ollama-handoff
Configuration Example
{
"remotes": [],
"packages": [
{
"registryType": "pypi",
"identifier": "ollama-handoff",
"version": "0.1.2",
"transport": {
"type": "stdio"
},
"environmentVariables": [
{
"description": "Base URL of the Ollama server.",
"default": "http://localhost:11434",
"name": "OLLAMA_URL"
},
{
"description": "Default model used for handoffs.",
"default": "qwen2.5-coder:14b",
"name": "OLLAMA_DEFAULT_MODEL"
},
{
"description": "Context window in tokens.",
"default": "32768",
"name": "OLLAMA_NUM_CTX"
},
{
"description": "How long to keep the model resident in VRAM.",
"default": "30m",
"name": "OLLAMA_KEEP_ALIVE"
},
{
"description": "Per-request timeout in seconds.",
"default": "600",
"name": "OLLAMA_TIMEOUT_S"
}
]
}
]
}
mcp
model-context-protocol
pypi
By
Comments
Sign in to leave a comment