Automated Learning Progress Tracker with Performance Analytics

Design a comprehensive Python script using pandas and scikit-learn that automatically aggregates student performance data from multiple learning management systems (LMS), calculates individual learning trajectories, and generates predictive analytics on student risk factors. The script should integrate data from Canvas, Moodle, and custom school databases, create visualization dashboards using Plotly, and output a JSON report with machine learning-powered intervention recommendations. Include error handling for disparate data formats and implement secure credential management.