The Profanity API
API for profanity detection that handles evasions, analyzes context and intent, and returns structured moderation scores
An API for profanity and toxicity detection that identifies evasive spellings, distinguishes intent and context, and returns structured scores to moderate user-generated text. It is for developers and engineering and product teams at companies building chat, gaming, and community platforms that need to filter content. It is positioned as a layered pipeline combining regex, vectors, and LLM analysis only when needed to balance speed, accuracy, and cost versus single-method filters.
Key features
- 5-layer profanity detection
- 12 context types
- 13 intent categories
- Adaptive latency modes
- Advanced normalization for evasions
- Homoglyph and unicode detection
- Phonetic spelling detection
- Layer scores with term positions
- LLM reasoning and debug logs
- Interactive playground
- Real-time JSON preview
- One-click starter presets
- Multi-language code snippets
- Edge-native global deployment
- No social media activity within the last 30 days
GTM channels
- API
- Docs
ICP
- Software developers
- Engineering teams
- Product teams