User Profile Moderation

SDK v7.x · Last verified March 2026 · iOS · Android · Web · Flutter

Speed run — just the code

// 1. Flag a user profile
await UserRepository.flagUser('userId', 'inappropriate_profile');

// 2. Unflag a user profile
await UserRepository.unflagUser('userId');

// 3. Update a user profile (triggers AI scan)
await UserRepository.updateUser('userId', {
  displayName: 'New Display Name',
  description: 'Updated bio text',
});

Full walkthrough below ↓

Platform note — code samples below use TypeScript. Every method has an equivalent in the iOS (Swift), Android (Kotlin), and Flutter (Dart) SDKs — see the linked SDK reference in each step.

Content moderation catches bad posts — but what about bad profiles? Offensive display names, NSFW avatars, and scam-laden bios erode trust before a user even posts. This guide covers the full user profile moderation pipeline: AI-powered scanning, the admin review workflow, profile reset tools, and blocklists.

Prerequisites: SDK installed and authenticated, Admin Console access for moderation configuration.Also recommended: Complete Content Moderation Pipeline first — user profile moderation extends the same AI and console infrastructure.

After completing this guide you’ll have:

User profile flagging via SDK (flag / unflag users)
AI scanning of display names, descriptions, and avatars
Admin review workflow with reset, clear, and ban actions, plus reset for any user from User History
Profile blocklist configured for display names and descriptions

How Profile Moderation Differs from Content Moderation

User profile moderation is flag-only — unlike posts and messages, flagged profiles are never auto-deleted. The block confidence threshold that auto-removes content does not apply to profiles. Admins must explicitly decide what to do.

Behavior	Content (Posts / Messages)	User Profiles
Pre-moderation	✅ Images blocked before upload	✅ Avatar images blocked before upload
Post-moderation	✅ AI scans after publish	✅ AI scans after profile save
Auto-delete at block confidence	✅ Content removed automatically	❌ Never — flag only
Admin reset	N/A	✅ Tick which fields to reset — flagged profiles from the Users tab, any user from User History

Flag-only for profiles: Flagged profiles always require manual admin review. This prevents automated systems from silently wiping user accounts.

Step-by-Step Implementation

Flag a user profile from the SDK

Let users report profiles they find inappropriate. Flagged profiles enter the moderation queue alongside AI-flagged profiles.

TypeScript

import { UserRepository } from '@amityco/ts-sdk';

// Flag a user with a reason
await UserRepository.flagUser('user-456');

// Unflag if the report was a mistake
await UserRepository.unflagUser('user-456');

Full reference → Content Flagging

AI scanning — what gets checked

When a user saves profile changes, AI scans three fields automatically:

Field	Scan Type
Display Name	Text
Description	Text
Avatar (uploaded file)	Image

Text is scanned for: harassment, sexual content, violence, hate speech, scam promotion, self-harm, and PII (URLs, person types).Avatars are scanned for: nudity, suggestive imagery, violence, hate symbols, and substance-related content.

External avatar URLs: Avatars set with avatarCustomUrl are not moderated. Only uploaded avatars (avatarFileId) are scanned, flagged, and reset.

Review flagged profiles in the Admin Console

Flagged profiles appear in Admin Console → Moderation → Moderation Feed → Users tab.Each flagged profile shows:

AI moderation labels with category counts (e.g., “Hate (2)”, “Violence (1)”)
Which field triggered the flag — display name or description
User report count from community members
Last flagged timestamp

Available actions:

Reset profile — restore flagged fields to safe defaults
Ban globally — ban the user across all communities
Clear flag — dismiss the flag and approve the profile

→ AI User Profile Moderation

Reset a profile

Whichever route you take, fields are restored to these defaults:

Field	Reset behavior
Display Name	Set to `{prefix}{UUID}` (e.g., `User_a1b2c3d4e5f6`)
Description	Cleared to empty
Avatar	Removed — reverts to default; original file deleted from storage

There are two places to reset from, both opening the same reset modal where you tick the fields to restore:

Where	Applies to
Moderation Feed → Users tab	Profiles in the review queue (AI-flagged and user-reported)
User History page	Any user, flagged or not

To reset a profile that was never flagged — or to reset just one field, such as an avatar while leaving the display name alone — open the user’s User History page, click the three-dot (⋮) menu, and choose Profile moderation → Reset profile. Tick the fields to restore; unticked fields are left untouched.The user receives a USER_PROFILE_RESET notification through the notification tray listing the fields that were reset.

Avatar reset is irreversible — the original image file is deleted from storage. Only the reset event is logged; the original image cannot be recovered.

Configure the profile blocklist

Block specific words and phrases from appearing in display names and descriptions.In Admin Console → Moderation → Blocklist, add terms under the User Profile category. This blocklist is independent from the content blocklist — you can maintain different blocked terms for profiles versus posts.

Blocklist Category	Applies to
Content	Posts, comments, messages
User Profile	Display names, descriptions

Admin Console: Moderation Feed — Users Tab

The Users tab in the Moderation Feed is the central hub for profile moderation.

To Review
Reviewed

Displays all profiles requiring moderator attention — both AI-flagged and user-reported.

AI moderation labels with detected categories and occurrence counts
(Profile description is flagged) label when the description triggered the flag
User report counts from other community members
Actions: Reset profile, Ban globally, Clear flag

Confidence Thresholds

Profile moderation reuses the same confidence thresholds configured for content moderation — but only the flag threshold applies.

Threshold	Applies to profiles?	Behavior
Flag Confidence (default: 40)	✅ Yes	Profile flagged for admin review
Block Confidence (default: 80)	❌ No	Does not auto-delete profile fields

Start with medium flag confidence (40–60) and adjust based on false positive rates. Since profiles are never auto-deleted, a lower threshold is safer — it just means more items in the review queue.

Webhook Automation

Profile moderation events trigger webhooks for downstream automation — sending notifications, syncing with CRM systems, or logging incidents.

Node.js

const express = require('express');
const crypto = require('crypto');
const app = express();

app.use(express.json());

app.post('/webhook', (req, res) => {
  const signature = req.headers['x-amity-signature'];
  const payload = JSON.stringify(req.body);
  const secret = process.env.WEBHOOK_SECRET;

  const expectedSig = crypto
    .createHmac('sha256', secret)
    .update(payload)
    .digest('hex');

  if (!crypto.timingSafeEqual(
    Buffer.from(signature, 'hex'),
    Buffer.from(expectedSig, 'hex')
  )) {
    return res.status(401).json({ error: 'Invalid signature' });
  }

  const { event, data } = req.body;

  switch (event) {
    case 'user.didFlag':
      notifyModerationTeam(data);
      break;
    case 'user.didBan':
      revokeAccessAcrossSystems(data.userId);
      break;
  }

  res.status(200).json({ received: true });
});

→ Webhook Events Reference

Common Mistakes

Relying only on AI scanning without a blocklist — AI moderation catches broad categories but may miss brand-specific terms, competitor names, or community-specific slurs. Always pair AI scanning with a curated blocklist for display names.

Assuming avatarCustomUrl avatars are moderated — Only uploaded avatars (avatarFileId) are scanned. If avatars need to be moderated, set them with avatarFileId.

Auto-banning on first flag — A single AI flag or user report may be a false positive. Use the reset action for first offenses and escalate to bans for repeat violations. Check user history for patterns before banning.

Best Practices

Graduated enforcement

Apply escalating actions based on violation history:

First offense — Reset only the offending field and notify the user
Second offense — Reset + temporary mute
Repeat violations — Global ban

Check the User History page for past moderation events before deciding on an action.

Blocklist maintenance

Review and update the profile blocklist monthly — slang and evasion tactics evolve
Add leetspeak and Unicode variants of blocked terms (e.g., h4te, ⓗⓐⓣⓔ)
Keep the profile blocklist separate from the content blocklist — profile terms are often different (impersonation names, brand abuse, etc.)

Team workflow

Assign profile moderation to a dedicated moderator or rotation — profile flags require different judgment than content flags
Use the Select moderator filter in the Reviewed tab for audit and load-balancing
Document your reset-vs-ban criteria so the team is consistent

User communication

The USER_PROFILE_RESET notification tells users what happened — but consider also linking to your community guidelines
Allow users to update their profile again after a reset — the new version will be re-scanned
Provide an appeal path for users who believe the reset was incorrect

Dive deeper: AI User Profile Moderation has the full console reference including detection categories, threshold configuration, and admin reset details.

Next Steps

Your next step → Roles, Permissions & Governance

Profile moderation is live — now configure who gets moderator access and what they can do.

Or explore related guides:

Content Moderation Pipeline

The full content moderation loop — AI screening, user flagging, admin review, and webhook automation

User Profiles & Social Graph

Build the profile pages and social graph that this guide moderates

Notifications & Engagement

Configure the notification tray that delivers profile reset alerts

Overview

Build Your Community

Create & Share Content

Communication

Engage & Grow

Livestream

Monetize

User Profile Moderation

How Profile Moderation Differs from Content Moderation

Step-by-Step Implementation

Admin Console: Moderation Feed — Users Tab

Confidence Thresholds

Webhook Automation

Common Mistakes

Best Practices

Next Steps

Your next step → Roles, Permissions & Governance

Content Moderation Pipeline

User Profiles & Social Graph

Notifications & Engagement

​How Profile Moderation Differs from Content Moderation

​Step-by-Step Implementation

​Admin Console: Moderation Feed — Users Tab

​Confidence Thresholds

​Webhook Automation

​Common Mistakes

​Best Practices

​Next Steps

Your next step → Roles, Permissions & Governance

Content Moderation Pipeline

User Profiles & Social Graph

Notifications & Engagement

How Profile Moderation Differs from Content Moderation

Step-by-Step Implementation

Admin Console: Moderation Feed — Users Tab

Confidence Thresholds

Webhook Automation

Common Mistakes

Best Practices

Next Steps