Content Moderation by Use Case

Every surface has a different moderation problem. Gaming chat needs to allow trash talk; usernames need to defeat homoglyph evasion; medical forums can't flag anatomy. The Profanity API ships 12 context types so you tune strictness per surface instead of fighting one global word list.

Gaming Chat

Moderate in-game chat, lobbies, and party voice transcripts without nuking the trash talk that makes competitive play fun.

context: gaming mode: balanced

Creative Writing & Fiction

Let authors write villains, dialogue, and dark themes on storytelling, fanfic, and roleplay platforms — while still catching abuse aimed at real people.

context: creative mode: smart

Education Platforms

Moderate student forums, classroom tools, and edtech submissions while still allowing the difficult topics that real learning requires.

context: educational mode: balanced

Healthcare & Medical Apps

Moderate patient forums, telehealth chat, and symptom descriptions without flagging clinical and anatomical language as profanity.

context: medical mode: balanced

Legal Tech

Moderate legal platforms, case discussion, and document review where quoting offensive material is part of the job.

context: legal mode: balanced

Real-Time Chat & Messaging

Moderate DMs, group chats, and live messaging at conversational latency without holding up the send button.

context: chat mode: fast

Comment Sections

Moderate blog, news, and article comments where the baseline toxicity is high and targeted abuse is common.

context: comment mode: balanced

Reviews & Marketplaces

Moderate product reviews, business ratings, and marketplace feedback where frustrated customers vent but rarely cross into abuse.

context: review mode: balanced

Profile Bios

Moderate user bios, about sections, and profile descriptions — persistent public text that represents your platform.

context: bio mode: strict

Professional Networks

Moderate B2B platforms, recruiting tools, and workplace communities where the bar for professionalism is high.

context: professional mode: strict

Kids & Child-Safe Apps

Moderate children's apps, games, and learning tools with zero-tolerance filtering and grooming-aware intent detection.

context: child_safe mode: strict

Usernames & Handles

Validate usernames, display names, gamertags, and handles at sign-up — short, high-evasion text that's permanent and public.

context: username mode: strict

Looking for code in your stack?

Browse the framework and language integrations to get started in minutes.

View integrations →