textRay · red flags in texting

evidence-based relationship reads · iOS

red flags in texting, defined by evidence. not vibes.

the internet has a thousand doom lists that end in "trust your gut". your gut has been wrong before. here's what the actual patterns look like in an actual chat, and how a scan decides whether they're really there.

a red flag is a pattern, not a feeling

one snappy reply is a tuesday. the same move, every time it matters, is a pattern. textRay scans the conversation, both sides, and names what it finds calmly, with the receipts attached. "one-sided effort — 71% pattern match", quotes underneath, counterevidence one tap away. no sirens. no listicle. just what the messages show.

five patterns, and what each looks like in a chat

  • flipping it on you. you raise a problem, and somewhere between message three and message nine you're the one apologising. watch the reversal: your complaint goes in, their grievance comes out.
  • rewriting history. "that never happened." "you said you were fine with it." if the same argument keeps getting rewritten, we kept the original.
  • running hot and cold. paragraphs on tuesday, "haha yeah" by friday. warmth that arrives on a schedule you can't predict. the unpredictability is the mechanism, not a side effect.
  • one-sided effort. who starts the conversations. who suggests the plans. who circles back after a fight. when all three arrows point the same way, that's not a vibe. that's a statistic.
  • the switch-up. warm until you're invested, then cooler, then the little put-downs. memory smooths this out. a timeline doesn't.
themi never agreed to come to your mate's birthday thing
youmonday. 8pm. "count me in x"

exhibit A: monday, 20:14. still on file.

what makes a flag a flag

the scan doesn't collect gotchas. a finding has to earn its place:

a finding has to earn it
  • repetition: the same move on different days, not one bad night
  • density: how much of the conversation the pattern actually takes up
  • direction: whether it flows one way or gets traded both ways
  • counterevidence: the apology that did arrive and the plan they did keep are counted, not hidden

one bad day is not a flag

everyone texts badly sometimes. tired, stressed, three drinks in, mid-deadline. a scan that flagged every off moment would be a horoscope with extra steps. so the read is hedged where the evidence is thin, and if the receipts can't back a claim, the claim doesn't ship. sometimes the report comes back "friction, then actual repair" and the red flag you were braced for isn't there. verdict: pending. evidence: available.

the receipts will decide

stop re-reading it. read it once, properly.

download on the App Store

iOS · 16+ · no subscription · credits from £2.99

the scan reads your side too

red flags aren't a team sport. the scan reads both sides of the conversation and follows the evidence wherever it goes, including towards you. sometimes the dry texter was you. sometimes the plans died in your court. the report says so, kindly, with the quotes to prove it. that's the point of a second opinion: it isn't on anyone's side.

The serious bit, read it anyway.

Some things are not patterns to debate. If a conversation includes threats, sexual pressure, or someone controlling your money or who you can see, textRay stops the humour and moves to plain language and support options. Nothing is analysed until you agree to AI processing, reports stay in your history until you delete them, and findings describe patterns in a conversation, never a diagnosis of a person. textRay is for ages 16 and over.

questions, answered plainly

what counts as a red flag in texting?

a repeated pattern, not a single bad message. being blamed whenever you raise a problem, plans that never materialise, warmth that switches off once you're invested. textRay names these from the evidence in the conversation, messages cited.

can one bad message be a red flag?

usually not. most patterns only matter with repetition and density: how often, over how long, whether it flows one way. a few things are serious even once, and the scan treats those differently. one snappy reply on a bad day is not a finding.

does textRay tell me the other person is bad news?

no. findings describe patterns in a conversation, never verdicts on a person. every finding cites its messages, and counterevidence is shown rather than hidden. if the receipts cut against the theory, the report says so.

what happens if the messages include threats or pressure?

threats, sexual pressure, and controlling behaviour are not patterns to debate. when a scan finds them, textRay drops the humour and moves to plain language and support options. that part of the app is never funny on purpose.