README.md

---
title: Codex Extractor
emoji: 🐨
colorFrom: indigo
colorTo: pink
sdk: gradio
sdk_version: 6.14.0
python_version: '3.12'
app_file: app.py
pinned: false
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

## Smoke Signal Ops

- Diagnostic audit trail: `smoke_signal/manifest/DIAGNOSTIC_AUDIT_TRAIL.md`
- Bug report template: `smoke_signal/manifest/BUG_REPORT_TEMPLATE.md`

## OCR Runtime (Mistral OCR)

- PDF OCR is wired to Mistral OCR using model `mistral-ocr-latest`.
- Add the Hugging Face Space secret:
  - `MISTRAL_API_KEY`
- The extractor sends uploaded PDFs to Mistral OCR as base64 `document_url` payloads and preserves page-order output for fingerprint metrics.

## Fingerprint Data Boundary

- Final author fingerprint records store derived numeric/categorical metrics only.
- Source manuscript text is not stored in the final fingerprint library.