FOSYS: An AI-based Workflow Automation System for SCRUM and Software Development Teams

Modern software teams rely on Agile and SCRUM practices, yet gaps persist due to manual note-taking, inconsistent task tracking, and delayed updates. This paper presents an AI-powered workspace management system that automates meeting transcription, extracts actionable tasks, validates pull requests, and generates role-based dashboards. The system integrates large language models for task extraction, structured database mapping, and webhook-driven PR monitoring to create a unified workflow environment. By reducing manual effort and enabling real-time visibility, the solution strengthens accountability, improves task clarity, and enhances collaboration. Experimental evaluations demonstrate reliable task extraction and accurate PR–task synchronization, highlighting the potential of AI driven automation to improve software project efficiency.