requirements.txt

# ── Core pipeline ──────────────────────────────────────────────────────────
docling>=2.5.0
instructor>=1.3.0
groq>=0.8.0
pydantic>=2.7.0

# ── PII masking ────────────────────────────────────────────────────────────
presidio-analyzer>=2.2.354
presidio-anonymizer>=2.2.354
# spaCy model (download separately after install):
#   python -m spacy download en_core_web_lg
spacy>=3.7.0

# ── Utilities ──────────────────────────────────────────────────────────────
python-dotenv>=1.0.0     # load GROQ_API_KEY / GROQ_MODEL from .env
pyyaml>=6.0.0            # parse config/settings.yaml and config/prompts.yaml

# ── API server (Visual Audit UI) ───────────────────────────────────────────
fastapi>=0.111.0
uvicorn[standard]>=0.30.0
python-multipart>=0.0.9  # required by FastAPI for UploadFile

# ── Provenance fuzzy matching ──────────────────────────────────────────────
rapidfuzz>=3.9.0

# Demo fixture generation
reportlab>=4.2.0