Public-Record Watch

What this is and isn't

• Everything runs in your browser. The exports you paste are never uploaded — there is no server here to send them to.
• This tool never scrapes a register itself. You (or your own export process) supply both dated, customer-authorized snapshots; this only compares them.
• It does not create personal profiles, rank people, decide suitability, or claim that a person is compliant, safe, or eligible — it reports added, removed, and changed records only. That judgment is yours.
• Every record's source URL is carried into the report so a reviewer can open the original public record before acting.
• Only anonymous, aggregate usage counts (for example "a comparison was run") may be recorded locally — never register names, record ids, or record fields.

Common questions

Does this fetch the register for me?

No. This tool never makes a network request to a register. You supply both snapshots — however you exported them — and this only compares the two. That keeps it usable for registers with no public API, a login wall, or a rate limit a general fetcher could never respect.

What counts as a valid snapshot?

A JSON object with authorized: true (you attest you're authorized to hold this export), a register name, a source_url the report links back to, and a records list where every record has a unique id. All other record fields are compared as-is.

Why doesn't a removed record mean the person lost their license?

A record missing from the "after" export could mean the license lapsed, the export was incomplete, or the register reorganized its ids. This tool reports the observation only; the reviewer opens the source URL and confirms what actually happened before relying on it.

Is my export uploaded anywhere?

No. Comparison happens with JavaScript running in your own browser tab. This page makes no network request with your pasted content — there is nothing to intercept because nothing is sent. Only anonymous, aggregate usage counts like "a comparison was run" may be recorded locally, with an opt-out below.

Is there a command-line version?

Yes — the same comparison logic ships as a local Python CLI (src/public_record_watch/cli.py) in this tool's repository, for scripted use. See the README for its usage.