Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for community moderators

Community Guideline Update Agent

Guidelines that match how the team actually moderates

CreatingCustomer support
See how it worksGet access
Community Guideline Update Agent: what goes in, what the agent does and what you get

What it does

Community guidelines go stale, and moderators end up making calls the rules never covered. Each quarter this agent reviews cases where moderators wrote not covered or disagreed, plus appeals that were reversed. It groups them into gaps in the guidelines. For each gap it drafts new or clearer wording with examples. Then it tests the draft on the past cases: would the new wording have produced the decision the team agreed on? If not, it revises the wording and tests again. It drafts a member-facing change note in plain language. The moderation lead approves every change and the announcement. Edge case: a change that would make many past actions wrong is flagged, so the team can decide whether to review those actions.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Quarterly guideline review 2 USES A TOOL Collect 'not covered' cases, disagreements andreversed appeals 3 DOES Group cases into guideline gaps 4 USES A TOOL Draft new wording with examples 5 CHECKS THE RESULT Would the new wording give the agreed outcome onpast cases? If not: revise the wording and retest. Back to step 4. 6 USES A TOOL Draft a plain-language change note for members 7 YOU APPROVE Lead approves the guideline change and announcement 8 RESULT Guidelines updated and moderators briefed
Read the steps as a list
  1. Quarterly guideline review
  2. Collect 'not covered' cases, disagreements and reversed appeals
  3. Group cases into guideline gaps
  4. Draft new wording with examples
  5. Would the new wording give the agreed outcome on past cases?If not: revise the wording and retest. Back to step 4.
  6. Draft a plain-language change note for members
  7. Lead approves the guideline change and announcementThe agent waits here for your OK.
  8. Guidelines updated and moderators briefed

How it decides

It ranks gaps by how often they came up and tests each draft on real past cases.

  • Gaps with five or more cases are addressed first
  • Every change includes an example
  • Changes that reverse many past actions are flagged

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Review frequency
  • Minimum cases per gap
  • Tone of member notes
  • Who approves changes

What keeps you in control

It always asks you first

  • Publishing guideline changes

Hard limits

  • Never publishes guideline changes
  • Never applies new rules to past cases on its own

It stops when

  • Done: changes approved
  • Stop: no gaps found this quarter

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensIn the Q3 review, 14 cases involved AI-generated images posted without labels. The first draft rule said such images must be labeled. Tested on past cases, it would have removed two memes the team had allowed, so the test failed. The agent added an exception for obvious parody and retested, and all 14 cases matched. The lead approved the rule and announcement.

More agents for community moderators

Risk management · Customer support

Spam and Scam Report Triage Agent

Clear scam and spam reports quickly while keeping false removals low.

11 steps2 checks2 jobs
Risk management · Customer support

New Member Welcome and Rules Check Agent

Reduce rule breaks from new members through timely, friendly guidance.

11 steps2 checks3 jobs
Risk management · Monitoring

Raid and Coordinated Abuse Response Agent

Contain a raid within minutes and prepare an accurate ban list.

11 steps2 checks1 job
Planning · Scheduling

Moderator Wellbeing Rotation Agent

Spread harmful content exposure fairly while keeping every queue covered.

11 steps2 checks1 job