# Probability and the gate

Use probabilities to route review, with an explicit risk policy.

`probabilities` describes the answer distribution. `probability` is the calibrated probability of the selected answer, `probabilities[choice]`. It is not an independent verifier or a separately learned certainty score, and it is not a normalized "confidence" that rescales by the number of options: `probability: 0.9` means 0.9.

## Server-side routing

Send `threshold` between `0` and `1`. The default is `0.9`, unless the service operator configures a different default. The effective value is returned in both `threshold` and `gate.threshold`.

- `probability >= threshold` produces `action: "act"`.
- `probability < threshold` produces `action: "ask_human"`.

Equality passes. The gate is a routing policy; it does not execute a refund, invoke a tool, or contact a reviewer for you. Your backend must honor `action` and enforce authorization and business constraints.

A threshold is not a guaranteed error bound. Calibrated probabilities describe frequencies over comparable decisions and can drift when data, prompts, rubrics, or model versions change. Validate the threshold against the cost of errors in your workflow.

## An illustrative response

```json response
{
  "id": "dec_illustrative",
  "model": "ren-1.1.0",
  "type": "choice",
  "choice": "billing",
  "probability": 0.82,
  "probabilities": {"billing": 0.82, "delivery": 0.18},
  "action": "ask_human",
  "threshold": 0.9,
  "gate": {"threshold": 0.9},
  "usage": {"input_tokens": 96, "output_tokens": 0, "model": "ren-1.1.0"}
}
```

These values are invented to show the wire format. They are not measurements. A higher threshold generally sends more work to review; it cannot repair missing categories or misleading evidence.

For true/false decisions, `probability` describes the winning answer, including false; the probability of true is `probabilities.true`. For scores, it describes the winning level, not the expected score.