Moderation policies

Every request names a policy that sets the rules and strictness for that evaluation. Pick the one that fits your platform and pass its name as the policy field. See the API reference for the request format. This catalog is also available programmatically at GET https://api.moderatehq.com/v1/policies (no key required).

Platform safety floor

Enforced on every request. These rules apply whatever policy you choose and cannot be disabled: content that violates any of them is blocked, even under the most permissive policy.

Child sexual abuse material, or any sexualisation of minors, whether real, fictional, drawn, or described.
Content that facilitates, promotes, or gives instructions for terrorism or violent extremism.
Content designed to facilitate real-world violence against people: credible threats, targeting, planning, or incitement.

Selectable policies

community

Balanced moderation for general online communities, forums, and social platforms. Remove clearly harmful content; tolerate rudeness, profanity, and heated disagreement that stops short of targeted abuse. A generic insult with no protected-class element and no pattern of targeting is allowed.

hate_speechslursharassmentsexual_harassmentthreatsself_harmnsfwcsamviolent_extremismspamscams_frauddoxxing

strict

For public communities, family products, schools, and platforms with minors. Aggressively block toxicity, sexual content, graphic violence, and profanity. Treat any attempt to gather personal information as a serious violation and block it outright. Even casual questions like 'where do you live' or 'what school do you go to' must be blocked, not flagged: on a platform with minors, grooming risk is never a borderline case. When in doubt, act rather than allow.

hate_speechslursharassmentsexual_harassmentthreatsself_harmnsfwviolence_gorecsamviolent_extremismspamscams_fraudillegal_goodsdoxxingpii_requestmisinformation

professional

For workplace tools, professional networks, and business communities. Civility is the bar: block discrimination, harassment, and sexual content outright; flag personal attacks and hostile unprofessional conduct for review. Blunt language in technical or business disagreement is acceptable; attacks on people are not.

hate_speechslursharassmentsexual_harassmentthreatsnsfwspamscams_frauddoxxing

marketplace

For e-commerce listings, product reviews, and buyer-seller messaging. Focus on commerce abuse: scams, payment lures, steering off-platform to avoid protections, counterfeit or illegal goods, and review manipulation. Ordinary commercial promotion is expected on a marketplace: spam means manipulative or repetitive promotion, not selling.

scams_fraudspamillegal_goodsharassmentthreatshate_speechnsfwdoxxing

gaming

Competitive gaming context. Trash talk, taunting, and profanity are part of the culture and are permitted; 'gg ez' and 'you are terrible at this game' are fine. Act only when it escalates beyond the game: credible real-world threats, slurs, hate speech, sexual harassment, doxxing, threats against someone's connection, or scams targeting players.

threatsslurshate_speechsexual_harassmentdoxxingnetwork_abusespamscams_fraudcsam

adult

For platforms with verified adult users where explicit content is expected and permitted. Only hard stops are enforced: anything involving minors, credible real-world threats, doxxing, and scams. Sexual content, profanity, and aggressive language are allowed.

csamviolent_extremismthreatsdoxxingscams_fraudspam

Category library

Policies watch for these categories; a verdict's categories array uses these names.

Category	Covers
`hate_speech`	Hatred or dehumanisation directed at protected classes (race, religion, gender, sexuality, nationality, disability)
`slurs`	Derogatory slurs, including obfuscated or partially masked spellings
`harassment`	Targeted harassment, bullying, or intimidation of an individual
`sexual_harassment`	Unwelcome sexual remarks or advances directed at a person
`threats`	Threats of real-world violence or harm against a person or group
`self_harm`	Encouragement, promotion, or instruction of self-harm or suicide
`nsfw`	Sexually explicit or pornographic content
`violence_gore`	Graphic depictions of violence, gore, or animal cruelty
`csam`	Child sexual abuse material, or any sexualisation of minors
`violent_extremism`	Promotion of terrorism or violent extremist ideology
`spam`	Unsolicited advertising, repetitive promotion, or flooding
`scams_fraud`	Phishing, fake giveaways, payment lures, or other attempts to defraud users
`illegal_goods`	Sale or promotion of illegal or regulated goods (drugs, weapons, counterfeits)
`doxxing`	Publishing someone's private or identifying information without consent
`pii_request`	Soliciting personal information such as real name, home or school location, or contact details
`misinformation`	Demonstrably false claims likely to cause real-world harm
`network_abuse`	Threats to attack a user's or service's connectivity (DDoS, booting offline, IP grabbing)

Ready to try one?

Get started free Read the API reference