com.atproto.moderation

​

Overview

​

Key Concepts

• Reports: User-submitted reports about problematic content or accounts
• Report Reasons: Categorized reasons for reporting (spam, harassment, etc.)
• Subjects: What can be reported (accounts, posts, profiles, etc.)
• Moderation Services: Where reports are sent for review

​

Procedures

​

createReport

await agent.com.atproto.moderation.createReport({
  reasonType: 'com.atproto.moderation.defs#reasonSpam',
  subject: {
    $type: 'com.atproto.admin.defs#repoRef',
    did: 'did:plc:spammer123'
  },
  reason: 'This account is posting spam links repeatedly'
})

await agent.com.atproto.moderation.createReport({
  reasonType: 'com.atproto.moderation.defs#reasonViolation',
  subject: {
    $type: 'com.atproto.repo.strongRef',
    uri: 'at://did:plc:user123/app.bsky.feed.post/3k2y...',
    cid: 'bafyrei...'
  },
  reason: 'This post contains harassment and threats'
})

curl -X POST https://bsky.social/xrpc/com.atproto.moderation.createReport \
  -H "Authorization: Bearer $ACCESS_JWT" \
  -H "Content-Type: application/json" \
  -d '{
    "reasonType": "com.atproto.moderation.defs#reasonSpam",
    "subject": {
      "$type": "com.atproto.admin.defs#repoRef",
      "did": "did:plc:spammer123"
    },
    "reason": "Posting spam links"
  }'

​

Type Definitions

​

reasonType

​

Legacy Reasons (Deprecated but still supported)

​

Modern Reasons (via tools.ozone.report.defs)

• tools.ozone.report.defs#reasonViolenceAnimal - Animal cruelty
• tools.ozone.report.defs#reasonViolenceThreats - Threats of violence
• tools.ozone.report.defs#reasonViolenceGraphicContent - Graphic violent content
• tools.ozone.report.defs#reasonViolenceGlorification - Glorification of violence
• tools.ozone.report.defs#reasonViolenceExtremistContent - Extremist content
• tools.ozone.report.defs#reasonViolenceTrafficking - Human trafficking
• tools.ozone.report.defs#reasonViolenceOther - Other violence

• tools.ozone.report.defs#reasonSexualAbuseContent - Sexual abuse content
• tools.ozone.report.defs#reasonSexualNCII - Non-consensual intimate imagery
• tools.ozone.report.defs#reasonSexualDeepfake - Sexual deepfakes
• tools.ozone.report.defs#reasonSexualAnimal - Sexual content involving animals
• tools.ozone.report.defs#reasonSexualUnlabeled - Unlabeled sexual content
• tools.ozone.report.defs#reasonSexualOther - Other sexual content issues

• tools.ozone.report.defs#reasonChildSafetyCSAM - Child sexual abuse material
• tools.ozone.report.defs#reasonChildSafetyGroom - Grooming behavior
• tools.ozone.report.defs#reasonChildSafetyPrivacy - Child privacy violation
• tools.ozone.report.defs#reasonChildSafetyHarassment - Harassment of children
• tools.ozone.report.defs#reasonChildSafetyOther - Other child safety issues

• tools.ozone.report.defs#reasonHarassmentTroll - Trolling
• tools.ozone.report.defs#reasonHarassmentTargeted - Targeted harassment
• tools.ozone.report.defs#reasonHarassmentHateSpeech - Hate speech
• tools.ozone.report.defs#reasonHarassmentDoxxing - Doxxing
• tools.ozone.report.defs#reasonHarassmentOther - Other harassment

• tools.ozone.report.defs#reasonMisleadingBot - Undisclosed bot account
• tools.ozone.report.defs#reasonMisleadingImpersonation - Impersonation
• tools.ozone.report.defs#reasonMisleadingSpam - Spam
• tools.ozone.report.defs#reasonMisleadingScam - Scam
• tools.ozone.report.defs#reasonMisleadingElections - Election misinformation
• tools.ozone.report.defs#reasonMisleadingOther - Other misleading content

• tools.ozone.report.defs#reasonRuleSiteSecurity - Site security threat
• tools.ozone.report.defs#reasonRuleProhibitedSales - Prohibited sales
• tools.ozone.report.defs#reasonRuleBanEvasion - Ban evasion
• tools.ozone.report.defs#reasonRuleOther - Other rule violations

• tools.ozone.report.defs#reasonSelfHarmContent - Self-harm content
• tools.ozone.report.defs#reasonSelfHarmED - Eating disorder content
• tools.ozone.report.defs#reasonSelfHarmStunts - Dangerous stunts
• tools.ozone.report.defs#reasonSelfHarmSubstances - Substance abuse
• tools.ozone.report.defs#reasonSelfHarmOther - Other self-harm content

​

subjectType

• account: User account (DID)
• record: Repository record (post, profile, etc.)
• chat: Chat message or conversation

​

Report Subjects

​

Reporting an Account

{
  $type: 'com.atproto.admin.defs#repoRef',
  did: 'did:plc:user123'
}

​

Reporting a Record (Post, Profile, etc.)

{
  $type: 'com.atproto.repo.strongRef',
  uri: 'at://did:plc:user123/app.bsky.feed.post/3k2y...',
  cid: 'bafyrei...'
}

​

Reporting a Blob (Image, Video)

{
  $type: 'com.atproto.admin.defs#repoBlobRef',
  did: 'did:plc:user123',
  cid: 'bafyrei...',
  recordUri: 'at://did:plc:user123/app.bsky.feed.post/3k2y...'
}

​

Common Use Cases

​

Report Spam Account

await agent.com.atproto.moderation.createReport({
  reasonType: 'tools.ozone.report.defs#reasonMisleadingSpam',
  subject: {
    $type: 'com.atproto.admin.defs#repoRef',
    did: 'did:plc:spammer123'
  },
  reason: 'Account is mass-following and posting promotional content'
})

​

Report Harassment

await agent.com.atproto.moderation.createReport({
  reasonType: 'tools.ozone.report.defs#reasonHarassmentTargeted',
  subject: {
    $type: 'com.atproto.repo.strongRef',
    uri: 'at://did:plc:user123/app.bsky.feed.post/3k2y...',
    cid: 'bafyrei...'
  },
  reason: 'This post contains targeted harassment and personal attacks'
})

​

Report Inappropriate Image

await agent.com.atproto.moderation.createReport({
  reasonType: 'tools.ozone.report.defs#reasonSexualUnlabeled',
  subject: {
    $type: 'com.atproto.admin.defs#repoBlobRef',
    did: 'did:plc:user123',
    cid: 'bafyreiblobcid...',
    recordUri: 'at://did:plc:user123/app.bsky.feed.post/3k2y...'
  },
  reason: 'Post contains sexual content without appropriate labels'
})

​

Report Impersonation

await agent.com.atproto.moderation.createReport({
  reasonType: 'tools.ozone.report.defs#reasonMisleadingImpersonation',
  subject: {
    $type: 'com.atproto.admin.defs#repoRef',
    did: 'did:plc:impersonator123'
  },
  reason: 'This account is impersonating @realuser.bsky.social'
})

​

Appeal a Moderation Action

await agent.com.atproto.moderation.createReport({
  reasonType: 'com.atproto.moderation.defs#reasonAppeal',
  subject: {
    $type: 'com.atproto.admin.defs#repoRef',
    did: agent.session.did // Your own account
  },
  reason: 'I believe my account was suspended in error. The flagged content was satire and did not violate community guidelines.'
})

​

Building a Report UI

function createReportForm(subject: { uri: string; cid: string }) {
  const reasons = [
    { 
      value: 'tools.ozone.report.defs#reasonMisleadingSpam',
      label: 'Spam'
    },
    { 
      value: 'tools.ozone.report.defs#reasonHarassmentTargeted',
      label: 'Harassment'
    },
    { 
      value: 'tools.ozone.report.defs#reasonSexualUnlabeled',
      label: 'Unlabeled Sexual Content'
    },
    { 
      value: 'tools.ozone.report.defs#reasonViolenceThreats',
      label: 'Threats or Violence'
    },
    {
      value: 'com.atproto.moderation.defs#reasonOther',
      label: 'Other'
    }
  ]

  return {
    reasons,
    async submit(reasonType: string, reason: string) {
      await agent.com.atproto.moderation.createReport({
        reasonType,
        reason,
        subject: {
          $type: 'com.atproto.repo.strongRef',
          uri: subject.uri,
          cid: subject.cid
        }
      })
    }
  }
}

​

Important Notes

​

Report Privacy

• Reports are submitted to the appropriate moderation service (PDS or labeler)
• The reported user is not notified who reported them
• Reports are reviewed by moderators, not automatically actioned

​

Report Limits

• Users can submit multiple reports
• Duplicate reports of the same subject are allowed
• Each report creates a new moderation queue item

​

Where Reports Go

• Reports for content/accounts on your PDS go to your PDS moderators
• Reports can also go to third-party labeling services if configured
• AppView (bsky.app) has its own reporting system for network-wide content

​

False Reports

• Submitting false or malicious reports may result in account suspension
• Use the appeal process if you believe action was taken in error

​

Related Lexicons

• com.atproto.admin - Administrative moderation actions
• com.atproto.label - Content labeling system
• tools.ozone.moderation - Ozone moderation service

​

Resources

• Moderation Guide
• Community Guidelines
• Labeling Specification