draft_report_summary
Build a bounded executive summary prompt from computed KPIs and verify cited figures against KPI data, flagging unverified numbers before report use.
Instructions
Build bounded executive summary prompt and verify cited KPI figures.
Args: computed_kpis: ComputeProcessKpisResult dict (or JSON string). audience: Target reader profile ("operations", "contract_admin", "external_client"). cadence: Reporting cadence ("daily", "weekly", "monthly"). candidate_narrative: Optional LLM-drafted prose to verify against KPI numbers. language: Prompt language ("es" or "en"). allowed_extra_numbers: Explicit whitelist for non-KPI reference figures (e.g. years).
Returns: Dict with bounded prompt bundle, verification flag, and any unverified numeric tokens.
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": {
"computed_kpis": "<computed_kpis>"
},
"name": "draft_report_summary"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| cadence | No | daily | |
| audience | No | operations | |
| language | No | es | |
| computed_kpis | Yes | ||
| candidate_narrative | No | ||
| allowed_extra_numbers | No |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||