What a Telegram Admin Bot Does for Your Group
A Telegram admin bot is an automated account you add to your group to do the moderation work an admin would otherwise do by hand: screen new members, enforce the rules, delete rule-breaking content, restrict repeat offenders, and report what happened. Telegram Bot App is built for exactly this. It watches every join and every message, acts within milliseconds of a violation, and gives you a web dashboard with the stats, the user reports, and a live feed of moderation actions. Think of it as the admin who never sleeps and never misses a message.
This guide explains what the bot handles for you, which jobs run on the free tier, and where the dashboard fits in. Every feature here is a toggle you turn on from the dashboard, not a command you have to memorise.
What the admin work actually is
Running a Telegram group by hand means doing the same things over and over: greeting joiners, checking whether a new account is a spam bot, removing scam links, muting someone who keeps breaking the rules, and keeping a record so your decisions stay consistent. The bot takes that list off your hands and does it the same way every time, day or night.
It does not replace your judgement. You set the rules, thresholds, and policy; the bot applies them consistently and you review the results. Where it acts automatically, auto-kicking a high-risk account or muting a repeat spammer, every action is logged so you keep oversight.
Member screening at the door
New members are where most trouble enters, so the bot screens them before they can post.
CAPTCHA on join challenges every new member with a simple task (a short arithmetic question or a button interaction) and keeps them read-only until they solve it. You configure a master toggle, a time limit from 1 to 60 minutes (15 is the default), and whether to kick members who do not verify in time. A script that joins hundreds of groups at once cannot solve an individual challenge for each one, so CAPTCHA breaks the automation at the join. Full setup is in the CAPTCHA verification guide.
Profile scanning runs alongside CAPTCHA. New members' profile pictures pass through NSFW image detection, and profile text is checked, so an account with an inappropriate avatar or bio gets flagged before it settles in. Both profile picture and profile text scanning are free-tier features.
CAPTCHA proves a member is human. It does not prove good intent, which is why the rest of the system keeps working after the join.
Rule enforcement on every message
Once a member is in, the bot reads each message against the rules you have enabled and acts on the ones that break them.
- AI Spam Intelligence scores each account on a behavioural risk scale from 0.0 (safe) to 1.0 (definitely spam), weighing violation history, profile signals, and behaviour across groups. When enabled, it auto-kicks anyone whose score reaches 0.75. Free for every group. See the AI Spam Intelligence guide.
- Spamfinder is a machine-learning classifier that scores message content for spam probability from 0% to 100%, threshold set by a slider (75% default). It deletes messages over your threshold and restricts the sender. Also free.
- NSFW image detection scans photos, GIFs, stickers, and profile pictures for pornographic, racy, or spoofed content, with an adjustable threshold. This is image classification, not text reading, and uses your image-scan quota.
- Sentiment and toxicity detection flags profanity, insults, and threats, each with its own punishment weight.
- Custom badwords adds your own banned terms, with matching that handles leetspeak and bypass attempts.
- Language enforcement removes messages not in your group's chosen language, across roughly 33 to 43 supported languages.
- File antivirus scans uploaded files.
- Invite-link and malicious-URL filtering blocks Telegram invite links (t.me/..., @channel) and flags malicious URLs through Google Safe Browsing.
A note on what this bot does not do: it blocks invite-link spam, but it does not add or mass-invite members, which violates Telegram's terms of service. It is a moderation bot, not a mass-inviter, not a userbot, and not an AI chat assistant. No OCR, no AI-generated chat replies.
Punishments that escalate on their own
You do not ban each rule-breaker by hand. When a detection system flags a violation, the decision engine classifies it by severity and applies a timed restriction: roughly 30 minutes for pornographic content, 15 for racy content, 5 for spam or invite links, and 1 minute for profanity, badwords, or language violations.
The bot then escalates for repeat offenders. It tracks each user's cumulative punishment history and adds to the base duration based on past restrictions, so a first minor offence is a small, recoverable penalty while persistent rule-breaking climbs toward 15-30 minutes even for low-severity violations. The restriction is a timed mute: the user stays in the group, can read, and is automatically unrestricted when the time elapses, with no admin action needed. Group administrators are exempt; their messages may be deleted and logged, but they are never restricted. The full model is in the automated punishment system guide.
This makes the system lenient with humans and firm with bots. A legitimate member who trips it once recovers quickly; a spam account that relies on posting volume piles up penalties fast, because posting more is what removes it.
The admin dashboard and analytics
A Telegram admin bot is only useful if you can see what it is doing. The web dashboard is where you configure every setting and review the results, from any device with a browser. You log in with Telegram, and it shows every group where you are an admin and the bot is present. Each group opens into three tabs:
- Info: subscription status, the administrator list (click an admin to see their intelligence report), and group metadata.
- Statistics: group metrics (total messages, total violations, punishment rate per 1,000 messages, last 7- and 30-day punishment trends, and a top-violations breakdown by type), plus the top 50 active members.
- Settings: every moderation toggle and threshold slider, grouped into Basic Protection, AI Moderation, Language, Messages & Greetings, and Advanced.
Three features make the dashboard a working analytics surface:
- User Intelligence reports show any user's spam rating (0.0-1.0), violation statistics, group activity, and full violation history. This is the same score the auto-kick uses, so you can check why an account was flagged.
- Live punishment feed streams moderation actions across all your groups in real time, each entry showing the user, violation type, confidence, and timestamp.
- Public spam-rating API returns a user's spam rating and risk level as JSON, with copy-paste examples in cURL, JavaScript, and Python.
For walkthroughs, see the dashboard guide and the User Intelligence analytics guide.
Welcome messages and rules
The bot also handles onboarding. From the Messages & Greetings settings you enable a custom welcome message for new joiners and a group-rules block, with an option to auto-generate the rules from your active settings. New members get the group's purpose and expectations the moment they arrive, without you posting a manual greeting each time.
What runs free, and what the paid plans add
The core moderation work is free. Every group gets AI Spam Intelligence, Spamfinder, CAPTCHA, invite-link blocking, custom badwords, language enforcement, the escalating punishment system, profile scanning, and the dashboard at no cost. The free tier also includes 500 image scans and 1,000 sentiment checks per month.
Paid plans add capacity and extra media moderation on top of that free core:
- Gold: $4.99/month
- Platinum: $9.99/month
- Ultimate: $49.99/month
Annual billing is offered at about 20% off. Compare what each tier adds in the premium guide.
Frequently asked questions
What does a Telegram admin bot do?
A Telegram admin bot automates the moderation work of a group admin: it screens new members with CAPTCHA and profile checks, reads every message against the rules you set, deletes rule-breaking content, restricts repeat offenders with escalating timed mutes, and records each action in a web dashboard. It runs automatically once enabled, so enforcement continues even when no human admin is online.
Is the admin bot free to use?
The core moderation runs free for every group: AI Spam Intelligence, Spamfinder, CAPTCHA, invite-link blocking, badwords, language enforcement, the punishment system, profile scanning, and the dashboard. The free tier also includes 500 image scans and 1,000 sentiment checks per month. Paid plans (Gold $4.99, Platinum $9.99, Ultimate $49.99 per month) add scan capacity and extra media moderation, about 20% off on annual billing.
Does the admin bot replace human admins?
No. You set the rules, thresholds, and policy; the bot applies them consistently and handles the routine volume. Automatic actions like the 0.75 auto-kick are logged in the live feed and User Intelligence reports, so admins keep oversight and can reverse a false positive. The bot frees admins from repetitive moderation; it does not make community decisions for them.
Can the admin bot invite or add members to my group?
No. It is a moderation bot, not a mass-inviter. It blocks invite-link spam (t.me/... links and @channel mentions) and detects malicious URLs through Google Safe Browsing, but it does not add or mass-invite members, because mass-inviting violates Telegram's terms of service. It also does not run OCR or post AI-generated chat replies.
How does the bot decide who to punish?
Each detection system (spam, NSFW, sentiment, badwords, language, invite links) reports a violation with a confidence score to a decision engine. The engine classifies it by severity, applies a timed restriction proportional to it, and escalates the duration based on the user's cumulative history. Group administrators are exempt. Every decision is logged, and you can open any user's intelligence report to see which factors raised their risk score.
What analytics does the admin dashboard show?
The dashboard shows per-group statistics (total messages, total violations, punishment rate per 1,000 messages, 7- and 30-day trends, and a top-violations breakdown), the top 50 active members, individual User Intelligence reports with each user's spam rating and full violation history, and a live punishment feed across all your groups. A public spam-rating API lets you pull a user's risk level into your own tools.
Getting started
To put an automated assistant on moderation in your group, add the bot and make yourself visible to the dashboard. Follow the step-by-step setup guide: add the bot to one group, grant it admin rights, and log in with Telegram to start configuring. The two strongest anti-spam layers are free, so a basic working setup costs nothing.