Scan free text for PII and report exactly where it is

A support ticket comes in with a customer's card number pasted right into the message, and nobody flags it before the model sees it.

IntermediateRead 6 minDoing: an afternoon8 steps

▶ Start

What you'll have at the end

A structured list of PII found in one document, each finding naming the type and where it sits

Before you start

You need

A working call to an LLM API that already returns JSON matching a schema you define, and one block of free text, like a support ticket, that you can read into a script as a plain string.

Not covered

Deciding which categories a law like GDPR or CCPA actually requires you to protect, and what to do with a finding once you have it, like redacting or blocking the message: this recipe only builds the report.

Leans on

8 steps

Where to go next

A hallucinated phone number breaks the record

you only need to confirm one already-extracted field is genuinely present in the source, not scan a whole document for several different sensitive types at once.

Checked 15 Aug 2026

Part of the Structured Output Generation with LLMs cookbook