govern_inference
Quality-govern an in-progress AI generation step BEFORE its output is used (complements validate, which checks finished documents). Returns an action - STOP, CONTINUE, REPAIR_REGION, REUSE_MOTIF, REVIEW, ESCALATE - with a plain-language explanation. Structural scores do not establish task correctness. Check safe_to_finalize and acceptance coverage. Persistence success is reported; read traces in the same Blueprint namespace.
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": {
"api_key": "<api_key>",
"task_type": "<task_type>",
"payload": "<payload>",
"inference_id": "<inference_id>"
},
"name": "govern_inference"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| source | No | Free-form caller label recorded for audit | mcp |
| api_key | Yes | GeodesicAI API key (gai_...) | |
| payload | Yes | Task-type payload: generative_text {text,...}; retrieval {query,candidates}; generic {features} | |
| blueprint | No | Owned Blueprint namespace for the trace | default |
| task_type | Yes | Kind of generation step being governed | |
| step_index | No | Step number within this generation (0-based) | |
| constraints | No | Optional governance constraint config object | |
| inference_id | Yes | Caller-chosen ID grouping the steps of one generation |