Welcome to Thresh

Thresh is a small, opinionated ETL service that takes spreadsheet exports of structured data and returns de-identified CSVs ready for analysis. It runs as an Express server with a POST /upload endpoint and a thin browser UI, and ships with a fully in-browser demo on GitHub Pages that processes files locally without any server round-trip.

The cleaning rules ship with healthcare-friendly defaults (NHI anonymization, DOB → Age, drop Address/Contact) but the underlying engine is general-purpose — the rules are a small, testable module you can extend or replace for any tabular-data cleaning workflow.

It is not a general-purpose data warehouse, a HIPAA-certified product, or a managed service — it is the cleaning layer that sits in front of one.

Try it without installing anything

A self-contained version of the cleaner — same rules, no server — runs entirely in your browser at /tool/. The workbook never leaves your machine.

What it does today

• Accepts .xlsx, .xls, and .ods uploads up to 5 MB.
• Walks every sheet in the workbook and applies a fixed set of cleaning rules:
  • Replaces the NHI column with sequential anonymized IDs (ID-001, ID-002, …) shared across all sheets in the same upload.
  • Converts DOB to a calculated Age.
  • Drops Address and Contact columns.
  • Removes empty rows, columns whose header looks like Column1, and exact-duplicate rows.
• Writes one CSV per sheet plus a JSON manifest summarizing what was processed.
• Optional: API-key auth, IP rate-limiting, and a virus-scan hook (ClamAV).

What it does not do

• No persistence beyond on-disk CSV outputs (which are swept after CLINISYNC_CSV_TTL_HOURS, default 24 h).
• No HIPAA controls: no audit log, no access roles, no encryption at rest.
• No .csv input or cloud-storage destinations.
• No CLI beyond a version banner — the API is the supported surface.

Where to go next

You want to…	Read
Try it in your browser	/tool/
Call the API	API - Upload Endpoint, API Examples
Understand the cleaning rules	Data Cleaning Rules, Field details
Run it locally	Project Scope, Technology
Understand the architecture	ETL Design, Security model