AiModeration

# AiModeration A PaperMC plugin that watches chat, batches messages, and asks an AI model to flag toxicity, spam, advertising, and grief threats — then applies the suggested punishment automatically. ## Features - Logs every chat message to `chat-msg.txt` in the plugin's data folder - Every 20 lines, the batch is flushed and sent to an AI provider for review - Supports OpenRouter and Groq as providers, with three model tiers (LIGHTWEIGHT, SMART, INTELLIGENT) - Applies `warn` / `kick` / `ban` actions straight from the console - Writes a permanent audit trail to `chat-action.log` - Optional Discord webhook integration — posts an embed for every action - `/aimoderation` command for reload, status, and manual batch sending ## Requirements - Java 21 - PaperMC 1.21.x - An API key for OpenRouter or Groq ## Building ``` mvn clean package ``` The compiled jar will be in `target/AiModeration-1.0.0.jar`. ## Installation 1. Drop the jar into your server's `plugins/` folder. 2. Start the server once to generate `config.yml`. 3. Fill in your `apiKey`, choose `apiProvider` and `aiIntelligence`. 4. Optionally enable the `discord` webhook section. 5. Restart or run `/aimoderation reload`. See [GUIDE.md](GUIDE.md) for full configuration and command details. ## Documentation - [GUIDE.md](GUIDE.md) — setup and usage guide - [SECURITY.md](SECURITY.md) — reporting vulnerabilities - [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md) — community guidelines - [release-1.0.0.md](release-1.0.0.md) — release notes ## License Provided as-is by CodeStuff.