scratch/data_audit.py

import os

superpacks = [
    "sentiaccounting", "sentianalysis", "sentibanking", "sentibiz", 
    "senticoach", "senticommunity", "senticorporate", "senticredit", 
    "sentiglobal", "sentiinsurance", "sentilaw", "sentiplan", 
    "sentirisk", "sentitax", "sentiwealth"
]

print("=== SUPERPACK DATA VOLUME AUDIT ===")
for sp in superpacks:
    path = os.path.join(sp, "cleaned")
    if os.path.exists(path):
        files = [f for f in os.listdir(path) if f.endswith(".json")]
        print(f"{sp:15} : {len(files):>5} records")
    else:
        print(f"{sp:15} : FOLDER MISSING")

print("\n=== KNOWLEDGE LIVE AUDIT ===")
live_path = "knowledge/live"
if os.path.exists(live_path):
    for root, dirs, files in os.walk(live_path):
        json_files = [f for f in files if f.endswith(".json")]
        if json_files:
            print(f"{root:30} : {len(json_files):>5} live records")
else:
    print("knowledge/live FOLDER MISSING")