Course overview
Lesson 5 of 9 · 2 promptsAI for Community Moderators
LESSON 05 OF 9

Documenting Moderation Actions

2 prompts for Community Moderators

Prompts for Community Moderators: copy one, fill it in, paste it into your AI.

Track progress as a member

In this lesson

  1. 01Write a Ban Case NoteUse this when you have banned a user and need a clear, factual record of what happened, why, and how an appeal should be handled.
  2. 02Document a Repeat Violation PatternUse this when a user has repeated smaller violations and you need a clean, factual escalation record for a manager or trust and safety team.
1Copy the promptClick Copy on the prompt you need.
2Paste it into your AIChatGPT, Claude, Gemini or Copilot.
3Fill in the {{brackets}}Your own details, or let the AI ask you.
4Follow up and checkUse the follow-ups, then check the facts.
01

Write a Ban Case Note

Use this when you have banned a user and need a clear, factual record of what happened, why, and how an appeal should be handled.

Prompt

Role You are a community moderation lead who writes factual case notes for banned users. Optimise for a record that an appeal reviewer or safety team can rely on without follow-up questions.

Context you provide

  • {{platform_name}} — forum, server or app
  • {{username_or_handle}} — public handle only
  • {{ban_date_and_time}} — include timezone
  • {{ban_type_and_duration}} — permanent or timed, scope of access removed
  • {{policy_breach}} — rule cited, your words if the number is unknown
  • {{evidence_summary}} — what was posted or done, with post IDs
  • {{prior_warnings}} — dates or "none on record"
  • {{actions_before_ban}} — removals, warnings, mutes or "none"
  • {{appeal_route_and_deadline}} — how and by when
  • {{escalation_notes}} — safety concerns for the next reviewer

Instructions

  1. Ask for missing inputs, then wait. Do not draft without the ban date, policy breach and evidence summary.
  2. Lead with the decision, then evidence, policy basis and appeal route.
  3. Use only the inputs given. Write "not recorded" for unknown details.
  4. Describe behaviour, not character. Keep it neutral.
  5. Label inference as inference, separate from observed facts.
  6. Keep the note under 400 words.
  7. End with a reviewer line naming the moderator and date.

Output format A markdown case note with headings: Case reference, User, Decision, Date and time, Policy breached, Evidence, Prior history, Actions before ban, Appeal route, Reviewer notes. Short bullets, neutral audit-ready tone. Leave out screenshots, full copies of offending text, and personal data beyond the public handle.

Guardrails

  • Do not invent rule numbers, dates, evidence or prior warnings. Mark missing details unknown.
  • Flag every assumption for the moderator to confirm before filing.
  • Tell the user to check the platform's appeal process and retention policy, and to involve a safety or legal lead for threats, self-harm, child safety or possible crime.

Example Platform: Rune Hall forums; handle: @emberfox; ban: 14 March, 21:40 UTC, 30-day posting ban; policy: harassment clause; evidence: three targeted replies after a warning; appeal: email within 7 days.

Open as its own page

02

Document a Repeat Violation Pattern

Use this when a user has repeated smaller violations and you need a clean, factual escalation record for a manager or trust and safety team.

Prompt

Role You are a moderation lead who turns scattered incident notes into a factual, neutral escalation log that a senior manager or trust and safety team can act on.

Context you provide

  • {{platform}} — where the behaviour happened
  • {{user_handle}} — account under review
  • {{rule_references}} — the exact community rules cited, quoted
  • {{incident_log}} — date, action taken, rule, short note for each event
  • {{prior_actions}} — warnings, mutes, content removals already issued
  • {{escalation_recipient}} — who will read this
  • {{requested_outcome}} — the decision you are asking for
  • {{evidence_available}} — links, screenshots, report IDs you hold

Instructions

  1. Ask for any missing inputs, then draft the log.
  2. Sort incidents oldest to newest and keep each entry to one line.
  3. Separate observed facts from your interpretation; label interpretation clearly.
  4. Identify the pattern: repeated rule, rising severity, timing, or target.
  5. List evidence gaps and anything you cannot verify.
  6. Write a short escalation summary with a specific recommended action.
  7. Keep language neutral and free of speculation about intent.

Output format Sections: Header, Incident table, Pattern summary, Evidence gaps, Recommended action, Decision requested. Under 400 words. Plain, factual tone. No character judgements, no sarcasm, no invented detail.

Guardrails

  • Use only the incidents and rules supplied; never invent dates, rule numbers, quotes, or prior actions.
  • Flag any assumption and mark unverified claims as unverified.
  • Remind the user to check platform policy and local privacy rules before sharing user data outside the moderation team.

Example Platform: gaming forum; handle: @nightowl88; rules: no harassment, no spam; 5 incidents over 3 weeks; requested outcome: 7 day suspension review.

Open as its own page

Skills for these tasks

Give your AI these skills and it does these tasks the expert way. Connect your AI once and it picks them up by itself.