The questions you were actually going to ask.

No, and it cannot. There is no part of ThreadSpot that writes to Reddit — no posting, commenting, voting or messaging. It never asks for your Reddit account and never holds one. You read the draft, change what you want, and post it yourself from your own account.

It would be, if it handed you a pitch to paste. Every draft is held to one rule: the first sentence has to give concrete help, and it must not mention your business. Drafts that break it are rewritten before you ever see them. The honest test is the one in the section above — would the reply still be worth reading if you deleted every mention of your product?

Nothing about ThreadSpot touches your account, so there is no automation for Reddit to detect. What can get you banned is how you reply, which is why most threads never reach you: of 65 candidates for one client recently, the model surfaced none, because none of them were conversations that wanted a supplier in them.

Only to post the reply, and only your own. ThreadSpot reads public posts through a third-party scraping service; it never signs in as you and never asks for a password or a session cookie.

It already has, twice, and that is precisely why this is built the way it is. Reddit closed unauthenticated access in May 2026 — plain scrapers now get a 403 regardless of headers. ThreadSpot reads through a paid service rather than pretending to be a browser, and sources sit behind one interface, so a second network was added without touching any of the matching or scoring. LinkedIn is already live.

Scraping is billed per result returned, at roughly $2 per thousand — a typical single-client run fetches a few hundred and costs well under a dollar. Reading each candidate with a model adds a fraction of a penny per post. The estimate sits on the button before you press it, and there is a hard per-run ceiling you set.