Back to Prompt Library
implementation
Build the application
Inspect the original prompt language first, then copy or adapt it once you know how it fits your workflow.
Linked challenge: Build a Wildfire Claim Evidence Ledger Reconciler and Claimant Explainer
Format
Code-aware
Lines
128
Sections
11
Linked challenge
Build a Wildfire Claim Evidence Ledger Reconciler and Claimant Explainer
Prompt source
Original prompt text with formatting preserved for inspection.
128 lines
11 sections
No variables
1 code block
## Build
Cross-reference asserted wildfire loss ledgers against burn severity maps located via STAC metadata. Generate reconciled loss summaries and clear claimant communication letters.
## User
Property Claim Adjuster Assistant
## Workflow
1. Parse claimant loss ledgers detailing destroyed outbuildings, perimeter fencing, and structural components.
2. Query public STAC catalogs to extract asset URLs for official fire containment and burn severity maps.
3. Correlate parcel boundary geometry against retrieved map assets to verify spatial impact.
4. Draft an itemized ledger reconciliation summary and a clear status letter explaining evidence status to the policyholder.
## Interface
Web portal showing parcel boundaries over satellite burn maps with an interactive claim ledger audit tool.
## Data
Query public STAC Item collections to locate map assets, using local mock files for automated unit testing.
Starter examples below are synthetic. They are not downloaded records.
## Source
[STAC specification](https://raw.githubusercontent.com/radiantearth/stac-spec/master/README.md)
Extracts asset URLs from STAC Item metadata to locate and retrieve physical maps for spatial containment tests.
## Tools
semantic-kernel, pystac, shapely, rasterio.
Integrates semantic memory and structured plugins to reconcile tabular ledger items against spatial fire map data.
Local or cloud LLM supporting structured function plugins and clear, empathetic writing.
## Deliverables
Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.
## Acceptance
- Evaluates whether claimant parcel geometry intersects retrieved map assets located via STAC metadata.
- Matches line-item loss entries with verified fire dates and flags incongruous expenses.
- Generates compassionate, factual claimant correspondence detailing necessary evidence for flagged items.
## Starter Examples
```json
[
{
"input": {
"claim_id": "CLM-FIRE-2024-91",
"parcel_coordinates": [
[
-121.5,
39.8
],
[
-121.4,
39.8
],
[
-121.4,
39.7
],
[
-121.5,
39.7
],
[
-121.5,
39.8
]
],
"stac_collection": "wildfire-maps",
"map_overlap_found": true,
"claimed_items": [
{
"item": "outbuilding_workshop",
"amount_usd": 45000,
"documented": true
},
{
"item": "irrigation_pvc_mains",
"amount_usd": 6200,
"documented": false
}
],
"note": "Simplified synthetic app record, not a complete official schema file."
},
"expected_output": {
"claim_id": "CLM-FIRE-2024-91",
"perimeter_match": true,
"verified_amount_usd": 45000,
"pending_amount_usd": 6200,
"claimant_letter": "Parcel spatial overlap with the Complex Fire confirmed via map assets. Workshop claim verified. Please submit photo or invoice documentation for irrigation PVC mains."
}
},
{
"input": {
"claim_id": "CLM-FIRE-2024-92",
"parcel_coordinates": [
[
-120.1,
38.2
],
[
-120,
38.2
],
[
-120,
38.1
],
[
-120.1,
38.1
],
[
-120.1,
38.2
]
],
"stac_collection": "wildfire-maps",
"map_overlap_found": false,
"claimed_items": [
{
"item": "timber_fencing",
"amount_usd": 8000,
"documented": true
}
],
"note": "Simplified synthetic app record, not a complete official schema file."
},
"expected_output": {
"claim_id": "CLM-FIRE-2024-92",
"perimeter_match": false,
"verified_amount_usd": 0,
"pending_amount_usd": 8000,
"claimant_letter": "Map assets located via STAC show no fire encroachment on your parcel. Please provide localized fire incident reports or field adjuster photos."
}
}
]
```
## Safety
Drafts reconciliation dossiers for adjuster review; does not issue binding claim approvals or reject legal coverage.
Use mock external writes. Require human review before any real deployment.Adaptation plan
Keep the source stable, then change the prompt in a predictable order so the next run is easier to evaluate.
Keep stable
Hold the task contract and output shape stable so generated implementations remain comparable.
Tune next
Update libraries, interfaces, and environment assumptions to match the stack you actually run.
Verify after
Test failure handling, edge cases, and any code paths that depend on hidden context or secrets.