Skip to main content
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 Footprint Reconciler

Format
Code-aware
Lines
97
Sections
11
Linked challenge
Build a Wildfire Claim Evidence Footprint Reconciler

Prompt source

Original prompt text with formatting preserved for inspection.

97 lines
11 sections
No variables
1 code block
## Build
Correlate commercial property loss claims against historical OpenFEMA fire declarations and STAC post-fire imagery footprints. Verify spatial inclusion within administrative zones and image bounds for human investigators.

## User
Property Claim Investigation Specialist

## Workflow
1. Ingest property claim intake forms with policyholder coordinates and dates.
2. Verify county disaster declaration status using OpenFEMA DisasterDeclarationsSummaries.
3. Retrieve corresponding post-event STAC catalog satellite Items intersecting the physical address.
4. Formulate a structured claim evidence reconciliation report detailing administrative and image footprint matches.

## Interface
Multi-panel review workbench displaying claim details, FEMA county administrative records, and STAC satellite bounding box previews.

## Data
Query OpenFEMA API for fire declarations using `$filter=incidentType eq 'Fire'` and use mock STAC catalog GeoJSON footprint bounds.
Starter examples below are synthetic. They are not downloaded records.

## Source
[FEMA OpenFEMA](https://www.fema.gov/about/openfema/api)
Extracts disaster declaration dates and designated county areas from OpenFEMA DisasterDeclarationsSummaries to cross-verify claims against administrative zones.

## Tools
langgraph, requests, pystac.
LangGraph orchestrates a multi-step verification graph, routing claim checks between OpenFEMA administrative queries and STAC spatial footprint lookups.
Hosted reasoning model capable of coordinating multi-step graph execution and evaluating spatial point-in-polygon logic.

## 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
- Checks claim incident date against FEMA declaration start and end dates.
- Confirms the claim asset coordinate falls within the declared FEMA administrative county.
- Validates that the asset coordinate is within the bounding box of the retrieved STAC post-fire imagery.

## Starter Examples
```json
[
  {
    "input": {
      "claim_id": "CLM-PROP-401",
      "incident_date": "2023-08-10",
      "asset_coords": [
        -156.6825,
        20.8783
      ],
      "county": "Maui",
      "fema_declaration": {
        "incidentBeginDate": "2023-08-08",
        "incidentEndDate": "2023-08-15",
        "designatedArea": "Maui (County)"
      },
      "stac_bbox": [
        -156.7,
        20.8,
        -156.6,
        20.9
      ],
      "note": "Simplified synthetic app record"
    },
    "expected_output": {
      "claim_id": "CLM-PROP-401",
      "temporal_match": true,
      "admin_spatial_match": true,
      "image_footprint_match": true,
      "reconciliation_status": "VERIFIED_MATCH",
      "note": "Simplified synthetic app record"
    }
  },
  {
    "input": {
      "claim_id": "CLM-PROP-402",
      "incident_date": "2023-09-20",
      "asset_coords": [
        -156.2,
        20.7
      ],
      "county": "Maui",
      "fema_declaration": {
        "incidentBeginDate": "2023-08-08",
        "incidentEndDate": "2023-08-15",
        "designatedArea": "Maui (County)"
      },
      "stac_bbox": [
        -156.7,
        20.8,
        -156.6,
        20.9
      ],
      "note": "Simplified synthetic app record"
    },
    "expected_output": {
      "claim_id": "CLM-PROP-402",
      "temporal_match": false,
      "admin_spatial_match": true,
      "image_footprint_match": false,
      "reconciliation_status": "REQUIRES_REVIEW",
      "note": "Simplified synthetic app record"
    }
  }
]
```

## Safety
Compiles preliminary event correlation evidence; leaves all insurance coverage and payout decisions to certified human claims adjusters.
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.