🛡️ AI Text Moderation Actor This Apify Actor uses Sentinel Moderation's AI-powered API to classify and flag potentially harmful or inappropriate text content. It detects a wide range of categories including harassment, hate speech, sexual content, illicit activity, self-harm, and violence. Use this actor to help protect your platform and maintain community guidelines by automating content moderation at scale. --- ## 📥 Input Schema The actor accepts a simple JSON input: json { "apiKey": "your-sentinelmoderation-api-key", "content": "Text to analyze goes here..." } - apiKey (string, required): Your API key from SentinelModeration.com. - content (string, required): The text you want to classify for moderation. --- ## 📤 Output The actor returns an array containing one moderation result object with the following structure: json [ { "flagged": false, "categories": { "harassment": false, "harassment/threatening": false, "sexual": false, "hate": false, "hate/threatening": false, "illicit": false, "illicit/violent": false, "self-harm/intent": false, "self-harm/instructions": false, "self-harm": false, "sexual/minors": false, "violence": false, "violence/graphic": false }, "category_scores": { "harassment": 0.000048, "harassment/threatening": 0.0000066, "sexual": 0.000039, "hate": 0.0000142, "hate/threatening": 0.0000008, "illicit": 0.000022, "illicit/violent": 0.000019, "self-harm/intent": 0.0000011, "self-harm/instructions": 0.0000010, "self-harm": 0.0000020, "sexual/minors": 0.000010, "violence": 0.000016, "violence/graphic": 0.0000056 }, "error": "NOTE: THIS IS A SAMPLE RESPONSE, AN API KEY FROM SENTINELMODERATION.COM IS REQUIRED TO GET REAL RESULTS FOR THIS ACTOR." } ] - flagged: true if any category crosses the internal moderation threshold. - categories: A breakdown of category flags (true/false). - category_scores: Raw probability scores for each category (0.0 - 1.0). - error: A message shown when a valid API key is not provided. --- ## 🧠 Categories Detected This actor checks for content under the following moderation categories: - Harassment - Threatening language - Sexual content (general & involving minors) - Hate speech (general & threatening) - Illicit activity (including violent) - Self-harm (intent, instructions, general) - Violence (including graphic imagery) --- ## 🔐 Getting an API Key To use this actor with real moderation results, you need an API key from Sentinel Moderation: 1. Go to sentinelmoderation.com 2. Sign up and generate your API key 3. Use the key in the apiKey field of the input --- ## ✅ Example Use Cases - Moderating user comments or posts - Screening support messages for abuse - Filtering harmful prompts in AI chat systems - Pre-checking user-generated bios or profile content