Spot bots before you waste time replying.
Every X reply in your inbox gets a bot-likelihood score. Three clear tiers (Human, Suspicious, Spam), so you can triage in seconds and ignore the noise.
great post, really resonated with what we're building
amazing thread!! check my bio for details 👀
curious how you handled the migration step?
DM me on Telegram for free signals: t.me/...
this is the part I keep coming back to too
Your inbox is full of replies that aren't worth your time.
If you reply on X with any consistency, you've felt it: a thoughtful post lands, the replies come in, and a third of them are reply-farm accounts, AI-generated filler, or telegram-link drops trying to ride your reach.
The hard part isn't ignoring obvious spam. It's the gray zone. The accounts that look real-ish. The replies that read fine until you check the bio. By the time you've audited a profile to decide whether to engage, the conversation has moved on.
BotBlock does the auditing for you. Every author in your inbox gets scored automatically against 30+ signals, and the result shows up as a small badge next to their handle. Real humans show no badge at all, so your inbox stays calm.
Score, tier, triage.
All happening in the background while you reply. No buttons to click.
Score
When a reply lands in your inbox, BotBlock pulls the author's last 20 posts and 40 replies and runs them against 30+ signals.
Tier
Scores collapse into three tiers: Human (no badge), Suspicious (amber), Spam (red). Hover any handle to see the top three signals that drove the score.
Triage
Flip on the “Hide bots” filter to drop everything Suspicious and Spam in one click. Your inbox is now humans only.
Built from years of seeing the same patterns.
No single signal flips an account into Spam. BotBlock requires several to converge, which keeps real humans clean while still catching the gray zone.
Replies in <30s on average
Behavior
Telegram or WhatsApp link in bio
Metadata
AI-generated text patterns
Language
Account less than 30 days old
Metadata
Posts duplicate content repeatedly
Behavior
Sycophantic opener + filler phrases
Language
Plus 24+ more signals across account metadata, posting behavior, and language patterns.
Included with every Pro plan.
BotBlock is part of ReplySocial Pro at $25/month. Unlimited scoring, no per-author fees.
Common questions.
Is BotBlock automatic?
Yes. As soon as a new post lands in your inbox, ReplySocial checks the author's account in the background. By the time you see the reply, you already know whether it's likely a real person, suspicious, or a bot.
What signals does BotBlock use?
Over 30 signals across three categories: account metadata (account age, follower ratio, blue check, bio content), behavior (reply timing, content duplication, posting cadence), and language (AI-generated phrasing, scam tier phrases, structural tells). Each signal contributes to a 0–10 score that maps to one of three tiers.
Will legitimate accounts get flagged as spam?
False positives are kept low because the algorithm requires multiple converging signals before tipping into Suspicious or Spam. New accounts are not flagged on age alone, and the most common case (real humans) renders no badge at all, so the inbox stays clean.
Can I disable BotBlock or change the threshold?
Badges are always shown so you can decide for yourself. The Hide bots filter is opt-in per account. There's no manual threshold knob, because most users want one click to hide spam, not five sliders to tune.
How is this different from Twitter's spam filter?
X's filter operates platform-wide and is intentionally conservative. It only catches the most obvious cases. BotBlock is tuned for power users replying in public: it surfaces the gray zone (rapid-reply farms, telegram-link bios, AI-generated answers) that X's filter consistently misses.
Stop wasting reply effort on bots.
BotBlock comes free with every ReplySocial Pro account. Start your free trial and see your inbox tonight.
