ContentAI: An AI-Powered Multi-Agent Social Media Content Automation Platform
B. Jahnavi
G. Anurupa
G. Komal Sai Kumar
G. Sumanth
Department of Artificial Intelligence & Data Science Vasireddy Venkatadri Institute of Technology (Autonomous)
Affiliated to JNTUK | NAAC 'A' Grade | Guntur, Andhra Pradesh - 522 508
Abstract — In today's digital landscape, maintaining an active and engaging social media presence is crucial for brand growth and visibility. However, continuous content creation demands significant time, creativity, and strategic effort, often overwhelming businesses and individual creators alike. This paper presents ContentAI, an AI- powered social media content automation platform that automates the end-to-end process of generating, previewing, editing, scheduling, and auto-publishing high-quality content across multiple social media platforms. The system employs a multi-agent AI architecture comprising five specialized agents — Input Analyzer, Strategy Architect, Content Writer, Image Creator, and Quality Reviewer — each powered by state-of-the-art large language models including OpenAI GPT-4o, Google Gemini 2.0 Flash, and Anthropic Claude
3.5 Sonnet. A Content DNA extraction engine captures each user's unique digital voice fingerprint and injects personalization into every generation request, ensuring all outputs reflect the user's established brand voice. The platform is built on a Next.js 16 full-stack architecture with Supabase PostgreSQL persistence and integrates AI image generation via DALL-E 3 and Stable Diffusion XL. Generated content is presented on an infinite React Flow canvas with 12 custom node types, providing real-time social media preview mockups for LinkedIn, Twitter/X, Instagram, and Facebook. An automated OAuth-based publishing engine publishes approved content to connected social accounts without any manual intervention. Experimental evaluation and pilot testing with 35 participants demonstrate that ContentAI reduces content creation time by over 85%, achieves a 97.3% auto-publish success rate across four major platforms, improves brand voice consistency by 50.2%, and scores 82.4 on the System Usability Scale, placing it in the 'Excellent' usability category.
Index Terms — Social media automation, multi-agent AI, large language models, content generation, brand personalization, Content DNA, React Flow, Next.js, Supabase, DALL-E, auto- publishing.
