AutoML is an MCP-powered platform that orchestrates end-to-end machine learning workflows, including data analysis, preprocessing, feature engineering, model selection, evaluation, and hyperparameter tuning. It offers data exploration capabilities such as computing dataset statistics (shape, memory usage, data types, missing values), efficient CSV loading with pandas and pyarrow, correlation analysis, and outlier detection. The preprocessing stage automates handling missing values, encoding categorical variables, and scaling numerical features, followed by feature engineering to prepare data for both regression and classification tasks. The platform supports a wide range of algorithms for regression (Linear Regression, Ridge, Lasso, ElasticNet, Random Forest, XGBoost, SVR, KNN, CatBoost) and classification (Logistic Regression, Ridge Classifier, Random Forest, XGBoost, SVM, KNN, Decision Tree, Naive Bayes, CatBoost). Evaluation includes regression metrics (R², MAE, MSE) and classification metrics (Accuracy, F1-Score), with confusion matrix visualization and model comparison. Hyperparameter tuning is automated with advanced search strategies, customizable scoring, and trial management. The project is organized into data samples, a tools module, utilities, and a server entry point (server.py) to run the MCP server, designed for integration with the MCP framework as described in the MCP docs.