• FLAGGED
  • CLEAN
  • REVIEW
  • LOGGED

How to automate content moderation with your AI

Connect OpenAI to your agent to catch policy violations. Once linked, your AI handles the scanning work for you.

Ask AI about this page

Short answer

Can my AI moderate content for me?

Yes. Your AI uses the OpenAI connector to check text against safety guidelines and flag violations. You'll see exactly what was flagged and why.

The results.

Where violations land.

The AI sorts every message into specific categories based on your safety rules.

FLAGGED

Policy violations

The AI identifies text that breaks your rules. You get a direct alert when a message is unsafe.

CLEAN

Safe messages

Messages that pass the check move forward without issue. The AI confirms they meet your standards.

REVIEW

Human oversight

Borderline cases are marked for your attention. You decide if the AI was too strict or too lenient.

LOGGED

Audit history

Every scan is recorded. You can review the history to see why certain content was flagged.

The workflow

What your AI does when text arrives.

Your agent acts as a filter, processing incoming text through a safety check.

  1. Ingest text

    Your agent pulls in messages from your connected channels or databases.

    moderate_content

  2. Scan for violations

    The AI checks the content against specific safety categories to find problematic text.

    moderate_content

  3. Label the content

    The AI assigns a category to the message, such as hate speech or harassment.

    moderate_content

  4. Report findings

    The AI tells you exactly which policy was triggered so you can take action.

    moderate_content

Try it

Copy these to start.

Use these prompts to trigger the moderation task.

Starting points

Replace the bracketed text with your specific channel names or policy types.

OpenAI Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_wqDLmxGMRbmUrKtHE8ZWJc73FuzFzNT70ItXxJkE/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — OpenAI capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "openai-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_wqDLmxGMRbmUrKtHE8ZWJc73FuzFzNT70ItXxJkE/mcp"
    }
  }
}
Copy into chat04

  • Moderate the last ten messages in my #general Slack channel.

  • Check this user comment for any hate speech violations.

  • Scan the incoming messages from my support queue against my safety policy.

  • Review the text in this Google Sheet and flag anything that violates our community guidelines.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect OpenAI once, then ask.

Link your account once. Your credentials stay encrypted, and you can start using the tools in any MCP-compatible client.

Connect OpenAI to your AI

FAQ

How the moderation behaves

  • 01

    Can the AI delete messages automatically?

    No. The AI only flags and categorizes content. You must manually delete or remove messages yourself.

  • 02

    Does the AI change the original text?

    No. The AI only reads and analyzes the text. It does not edit or rewrite the messages it scans.

  • 03

    Will it flag messages that are actually fine?

    Yes. The AI can be over-sensitive, so you should review flagged items to ensure accuracy.

  • 04

    Can I use this for private direct messages?

    Yes, provided your AI client has access to those messages through the connection.

  • 05

    Does it work with any text source?

    It works with any text you can feed into your AI client once the connection is active.

  • More questions about OpenAI? The Connector page answers them. See everything the OpenAI Connector can do

Connect OpenAI to Claude, Cursor, ChatGPT & more