Modly AI

Modly AI is a focused, AI-powered content moderation API that helps Web3 platforms review and flag inappropriate website content—specifically before publishing to IPFS. It analyzes both text and image links within HTML files to detect violations across key moderation categories.

Key Features:

• HTML Content Moderation Scans submitted website files for inappropriate or harmful text and image links, ensuring content quality before IPFS publishing.

• Category-Based Detection Flags content across currently supported categories:

  • Adult/NSFW Content

  • Violence

  • Hate Speech

  • Graphic or Disturbing Imagery

  • Harassment

  • Illegal Substances

  • Nudity

• Text & Image Analysis

  • Text is analyzed for tone, profanity, harassment, and other violations.

  • Image links (via <img> tags) are passed to vision models for nudity and graphic content detection.

• Severity Scoring & Insights Every flagged issue includes:

  • Category (e.g., harassment, profanity)

  • Severity level (low, medium, high)

  • Confidence score

  • A relevant excerpt or image URL

• Structured JSON Output The moderation response is cleanly formatted for easy frontend rendering or decision-making logic in your dApp.

Who It's For

Modly AI is built for:

• Web3 website builders

• IPFS-based publishing platforms

• DAO tools

• Community publishing interfaces

• Any dApp aiming to maintain ethical content standards before publishing permanently on-chain

Modly AI helps decentralized platforms uphold ethical publishing standards, reduce exposure to harmful content, and provide users with a safer, more trustworthy content experience.