Skip to main content
Loading navigation
Back to prompt library

Build a Wildfire Exposure Boundary Metadata Auditor — implementation

Linked challenge: Build a Wildfire Exposure Boundary Metadata Auditor

Prompt source

implementation61 lines11 sectionsNo variables1 code block
## Build
Evaluate commercial real estate exposure locations by compiling intersecting STAC metadata records. Identify available high-resolution multispectral imagery links for defensive vegetation buffer audits using an LLM-assisted metadata retrieval workflow.

## User
Commercial Property Risk Surveyor

## Workflow
1. Load commercial facility polygon geometries and perimeter buffer specifications.
2. Discover recent high-resolution STAC Items covering the extended perimeter zones.
3. Extract valid spatial asset links representing optical imagery bands from the STAC records.
4. Generate a structured metadata retrieval report for downstream vegetation analysis by human reviewers.

## Interface
Web inspection terminal presenting property boundary overlays and retrieved STAC metadata links.

## Data
Query open STAC Catalog collections for Sentinel-2 or NAIP metadata, or utilize bundled local STAC GeoJSON files.
Starter examples below are synthetic. They are not downloaded records.

## Source
[STAC specification](https://raw.githubusercontent.com/radiantearth/stac-spec/master/README.md)
Traverses STAC Collections and Item metadata to parse asset bounding boxes and multispectral imagery links.

## Tools
anthropic-claude, pystac, shapely.
Claude parses complex STAC JSON schemas to extract valid optical imagery links and formats structured retrieval summaries.
Hosted Claude API for contextual JSON parsing and metadata audit reporting.

## 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
- Queries STAC catalogs to find items intersecting property polygons.
- Successfully filters items containing optical imagery asset links.
- Emits human-readable metadata retrieval summaries for property inspection teams.

## Starter Examples
```json
[
  {
    "input": {
      "facility_id": "FAC-WUI-01",
      "buffer_clearance_m": 12.5,
      "stac_collection": "naip"
    },
    "expected_output": {
      "facility_id": "FAC-WUI-01",
      "imagery_links_found": 2,
      "metadata_status": "READY_FOR_REVIEW",
      "record_type": "simplified synthetic app record"
    }
  },
  {
    "input": {
      "facility_id": "FAC-WUI-02",
      "buffer_clearance_m": 45,
      "stac_collection": "naip"
    },
    "expected_output": {
      "facility_id": "FAC-WUI-02",
      "imagery_links_found": 0,
      "metadata_status": "NO_IMAGERY_FOUND",
      "record_type": "simplified synthetic app record"
    }
  }
]
```

## Safety
Provides preliminary metadata screening; does not execute physical site audits or replace qualified municipal fire marshals.
Use mock external writes. Require human review before any real deployment.