TraceReady public cocoa pilot
Reproducibility manifest

Dataset:
- Dataset title: Colombian-Cocoa-Dataset
- Dataset URL: https://www.kaggle.com/datasets/lehetasa/colombian-cocoa-dataset
- License reported by the source audit: CC BY-NC-SA 4.0
- Public rows analyzed: 57,658

TraceReady-supplied assumptions:
- country=Colombia
- commodity=cocoa

Method boundary:
- TraceReady used public row-level area and coordinate columns.
- TraceReady did not invent missing plot IDs, supplier identity, shipment linkage, or polygon geometry.
- No raw source rows or farm coordinates are redistributed in this pack.
- The derived output is a readiness report, issue summary, buyer handoff summary, source-owner follow-up list, and audit JSON.

Deterministic outputs:
- Records analyzed: 57,658
- Records with latitude/longitude: 57,658
- Records over 4 hectares: 46,134
- Polygon records present: 0
- Point-only plots over 4 hectares: 46,134
- Blockers found: 161,450
- Warnings found: 57,658
- Ready records: 0
- Readiness score: 0/100

Issue-code counts:
- missing_farmId: 57,658
- missing_supplier: 57,658
- missing_batch: 57,658
- polygon_required: 46,134

Trust boundary:
This public pilot is not a customer case, paid transaction, buyer approval, legal certification, audit assurance, TRACES submission, or due-diligence statement. It is a reproducible public-data stress test of TraceReady's file-readiness checks.
