Hey PH 👋 I built ZOHAL because I kept having this annoying problem: I wanted to use AI on real chats and docs… but the real stuff is always full of names, emails, company names, pricing, client context, and random details I didn’t want to paste anywhere. So the MVP is simple: **sanitize private chats and documents before using AI on them.** Handles PII in both English and Arabic You paste or upload a file, ZOHAL finds the sensitive bits, suggests safe aliases like `[PERSON_1]` or `[COMPANY_A]`, and lets you review them before creating a clean version. Then you can work with that sanitized version; ask questions, explain parts, check claims against the source, and export something safe to share. I’m calling it a sanitization workbench for now, but honestly I’m still figuring out the exact positioning. Would love feedback from anyone who deals with messy private docs, client notes, transcripts, or internal chats.

Zohal

An AI sanitization workbench for chats and documents.

An AI sanitization workbench for chats and documents.