run_ollama_task
Delegates heavy code, text generation, or analysis sub-tasks to a local or cloud Ollama model, reserving Claude tokens for work needing Claude-level reasoning.
Instructions
Delegates a heavy code, text generation, or analysis sub-task to a local or cloud Ollama model.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"prompt": "<prompt>"
},
"name": "run_ollama_task"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | The Ollama model tag to invoke. Use list_ollama_models to see what's pulled. | qwen-context:h-q4_0-24k |
| format | No | Constrain output: "json" for any JSON, or a JSON Schema object to force an exact shape. | |
| prompt | Yes | The specific task or prompt to send to the Ollama model. | |
| num_ctx | No | Context override; inherits the model setting when omitted. | |
| timeout_ms | No | Request deadline in milliseconds; defaults to the server's configured deadline (120 seconds by default). | |
| num_predict | No | Output ceiling override; defaults to the smaller of the saved model ceiling and 8192 tokens. | |
| system_prompt | No | Optional instructions framing the model's role. |