OnlyFans Restricted Words: Free Checker, Nine Categories and a Pre-Send Checklist

What are restricted words on OnlyFans?

Restricted words on OnlyFans, often called banned words, are terms that public guides report can get a message blocked, a caption rejected or an account flagged. Outseeker's guide says the filter reads every text field, from a creator's name and bio to captions, mass messages and DMs. Supercreator's guide says the terms mostly concern illegal activity, abuse, violence, anyone under 18 or non-consensual behavior.

The guides disagree on how many words there are: social-rise's guide says just over 200, XCreatorMgmt's headline promises 180+, and Supercreator's numbered list has 122 entries. The checker above therefore uses one rule: a word is in only when at least two of seven guides list it (social-rise, Supercreator, XCreatorMgmt, Infloww, Boese VA, Outseeker and OnlyMonster, read on 8 October 2026). That gives 184 words in nine categories, 24 of them context words such as dog, drink and meet, which are innocent in most sentences.

The checker marks a hard flag "to change" and a context word "check in context". It runs in the browser and sends nothing anywhere, and it finds whole words in any capitals, with an optional plural s, and a two-word term such as cash app written with a space or a hyphen.

Which categories of words are restricted on OnlyFans?

The checker sorts its 184 words into nine categories: age and youth, non-consent and incapacity, violence and extreme practices, family roles in a sexual context, animals, bodily fluids and functions, drugs and intoxication, meet-ups and sex work, and payments and platforms outside OnlyFans. Each category below gives the reason the guides report for it and at most three plain example words. The full list stays in the checker, which shows only the words found in the text you paste.

What happens when a message uses a restricted word?

According to the guides, a message with a restricted word can be blocked, silently dropped or deleted, and the account can be flagged, warned, suspended or banned. Outseeker's guide says a message may look sent on the sender's side while the fan never receives it. OnlyMonster's guide describes the same silent failure and says the platform sometimes refuses to send, with an error or an inactive send button. social-rise's guide says a message with a restricted word is deleted and never reaches the subscribers.

On the account side, Supercreator's guide says flagged content can be taken down and the account flagged, as quickly as within 24 hours, and XCreatorMgmt's guide also says a flag can come within 24 hours. Outseeker's guide says repeated use triggers a manual review and warning emails, with a temporary suspension for serious cases and a permanent ban for extreme or repeated ones. On appeals the guides disagree: social-rise's guide says accidental use will likely be reinstated, while Supercreator's guide says recovery is rare.

The guides also say the filter is more than a word list. OnlyMonster's guide says moderation looks at patterns and context, so the same phrase can pass in one setting and be flagged in another, and a word that is fine alone can look different beside a number and a verb that point to an arrangement off the platform. That is why the checker above reports words and leaves the reading of meaning to a person.

The restricted-words pre-send checklist (copy this)

Version 1.0, 8 October 2026. Copy the ten fields into the team's notes and fill one copy for every new caption, bio or mass-message template. A template reaches many fans at once, so write it with OnlyFans mass message templates and run it through this list before it is scheduled. Boese VA's guide says that for age, family, non-consent and animal topics there is no safe wording and a chatter should decline the topic instead of rephrasing, which field 6 asks you to decide. Keep every field, even when the answer is "none".

• 1. Where it goes: [DM, mass message, PPV caption, post caption or bio, and the creator]
• 2. Line (before): [the exact text as first written]
• 3. What it is for: [welcome, re-engage, answer a fan, sell a paid message, thank for a tip]
• 4. Checker result: [paste the line into the checker; write each flagged word and whether the checker marks it "to change" (a hard flag) or "check in context" (a soft flag)]
• 5. Category: [which of the nine categories each flag sits in, or "none"]
• 6. Rephrase or decline: [rephrase when the chatter wrote the word; decline in character and change the subject when the fan's request itself is about age, non-consent, family roles or animals]
• 7. Line (after): [the same purpose with no flagged word, or the decline]
• 8. Re-check: [the after line pasted into the checker again: clean, or what remains]
• 9. Meaning check: [read the after line as a stranger would: does it still point at one of the nine categories? yes or no]
• 10. Approved by and date (agency only): [the chatter, the manager who signed it off, the date]

Filled examples: six risky lines, rewritten

Each example fills the same ten fields in the same order. The lines are invented and non-explicit, and each checker result comes from running the line through the checker on this page. A clean re-check means no listed word remains, not that OnlyFans accepts the line.

How can an agency keep a whole team inside the rules?

Anlora's Message Guard, which we make, checks a message written by a person on your team before it is sent. It stops a send that contains an email address, a phone number, a link, a social handle, a crypto wallet or a word from the agency's own banned-word list of up to 200 words. Words match as whole words, so banning cash does not stop a message about cashmere. The list starts empty and the agency adds its own words. New teams start in block mode, which refuses the send; a warn mode only records what it would have stopped and sends anyway. The Guard keeps a log of what it stopped.

The Guard checks people, never AI replies, and it is a safety net for careless sends, not a guarantee. Anlora also chats with every paying fan on its own: see the OnlyFans AI chatbot page.

Other tools do similar work: Outseeker's guide says Infloww and OnlyMonster offer restricted-word detection that catches words before a message is sent, and Supercreator's help article says Supercreator alerts you when a message or post contains a restricted word so that it is not sent. Outseeker's guide also advises restricted-word training in onboarding and a weekly review of chat logs. The same habits work without a tool: keep a filled checklist with each template, have a manager approve new mass messages, and teach the nine categories in how to train OnlyFans chatters.

What mistakes make restricted-word problems worse?

• Checking only the DMs. The checker takes any text, so run bios, captions and mass-message templates through it as well, not only chat replies.
• Using spelling tricks. social-rise's guide says OnlyFans catches substitutions such as "M33T" in the end. XCreatorMgmt's guide says a discovered workaround can lead to permanent suspension, and Outseeker's guide says trying to get around the filter is worse than tripping it by accident.
• Leaving fan-driven topics to each chatter's judgment. Two of the common chatter mistakes Outseeker's guide lists are discussing meet-ups when a fan asks and naming payment apps when a fan suggests paying elsewhere. Agree one reply for each.
• Checking once and never again. Boese VA's guide says the list changes without notice, so re-run saved templates through the checker when the date on this page changes.
• Keeping no record. Without a filled checklist beside each template, nobody can see who approved a line or why a flag was kept.

What comes after the check?

Once a line passes, write the rest of the page copy with OnlyFans captions and OnlyFans bio examples, and run each new text through the checker before it goes live.