Prashna accuracy against a shuffled-time control
get_prashna_calibrationRead-only. For resolved Prashna questions (outcome recorded as confirmed, partly-confirmed or not-confirmed), reports per question area how often the real-time Yes/No verdict matched what happened, beside the same question recast at random times within 30 days of asking. Returns n, hit rate, control hit rate, their difference and an insufficient-data flag under 20 scored cases. Aggregates only: no question text or tokens.
Tool call template
This tool declares no required parameters. Add optional parameters from the input schema as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {},
"name": "get_prashna_calibration"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Most recent resolved questions to include | |
| category | No | Limit to one question area, e.g. career | |
| controlSamples | No | Random recastings per question for the control. Each is a full chart calculation, so at most 120 are run per call, newest questions first; the rest still count toward the hit rate (see controlledN). |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| schemaVersion | Yes |