What the game actually says
The premise is short. A sentient AI chip went rogue and took its hosts with it. Augs look like everyone else until they do not. Every faction agrees on shoot-on-sight, and you have the gun to enforce it. Shooting a human is logged as manslaughter, which is the sentence that makes the mechanic a decision rather than a chore.
The build ships six official tells: unusual pupils, an uncanny expression, a visible chip, unnatural stillness, twitching, and clothing that does not match their faction. Robotic or over-formal speech is a further signal that appears in the same material. Those are the tells the game gives you, not a fan list.
There is also a scanner: it reports only “AUG DETECTED” or “No Aug detected”, costs 25 power, and works once per customer. It is a second opinion, not an oracle. A guide that calls it 100 percent reliable is wrong about the one thing that matters.
The cheapest test in the game
Augs refuse every price negotiation. Not sometimes — every time. That single behaviour is the most reliable check available, and it costs nothing, because a normal customer will haggle and an Aug will not move.
Use it in order. Offer a price. If they haggle, they are human and the transaction continues. If they refuse to engage at all, you have a signal, and you can escalate to the scanner or to a visual check before deciding anything irreversible.
The value of this test is that it is procedural rather than perceptual. You do not need to read a face or judge a twitch; you need to notice an absence. Absences are easier to be consistent about, and consistency is what keeps a shooting policy from becoming a mood.
Why this is a trap with a margin
The decision is asymmetric. Shoot an Aug and you have done what the station asked, probably earning standing with everyone. Shoot a customer who was not an Aug and you have killed a paying customer in your own shop, with a manslaughter entry and a faction cost. Let an Aug walk and you have kept the sale, possibly kept your licence, and failed the one policy nobody disagrees about.
Most players resolve this by shooting rarely and early, or late and never. Both are coherent. The failure state is the middle: shooting sometimes, by mood, with no rule you could state afterwards.
There is a commercial angle the store page implies rather than states. Augs are customers, and a shop that serves them has a customer base nobody else will serve. That is real margin and real risk, and the game does not pretend otherwise.
A procedure you can defend
First, separate suspicion from evidence. Suspicion is a feeling about a customer; evidence is something you could describe to a security officer without adjectives. Second, prefer reversible moves: refusing service, delaying a sale, or reporting are all reversible in the sense that they do not end a life. Shooting is not.
Third, decide the shop’s policy before the first suspicious customer and write it as one sentence. “I shoot when I am certain” and “I never shoot” are both defensible. “I shoot when I am annoyed” is not a policy.
Fourth, if you do shoot, treat it as a transaction with a receipt. Note who saw, which faction gained, which lost, and whether the interrupted sale was worth more than the standing you gained. That habit is worth more than any tell list.
After the shot
The immediate aftermath is administrative: a customer is gone, a sale did not happen, and a line is written. The delayed aftermath is the one that matters. Factions that approve may reward you; factions that depended on that customer may not.
If you run a legal, high-reputation shop, an unnecessary shooting is the most expensive mistake available. It converts your best asset — the assumption that you are a legitimate business — into evidence. If you run a grey shop, the calculus is different and arguably easier: you already have a file.
Either way, do not shoot to recover a bad week. The policy is about Augs, not about stress.
Lower Levels Ledger. “How to spot an Aug, and when to shoot.” Last updated 2026-10-29. https://probablystolen.games/lore/augs/