Feature Scaler Engine MCP Connector for Claude
A+Standardize (Z-Score) or MinMax scale numeric columns with mathematical perfection local. Essential normalization for neural networks and clustering algorithms.
Neural Networks and K-Means clustering algorithms fail spectacularly if features aren't normalized. If an LLM attempts to subtract the mean and divide by the standard deviation across 5,000 rows, it will hallucinate 90% of the math.
This MCP brings deterministic Feature Scaling to your AI using simple-statistics. The AI specifies whether it wants Standard scaling (Mean=0, Variance=1) or MinMax scaling (Range 0-1), and the engine flawlessly transforms the target columns in milliseconds — returning the exact computed metrics for auditability.
The Superpowers
- Flawless Normalization: No LLM math hallucinations — exact scaling computed by your CPU.
- Multi-Column Support: Scale multiple features simultaneously in a single call.
- Automated Metric Extraction: Returns the exact Means, Std Devs, Mins, and Maxs used for scaling.
- Data Privacy: Your sensitive training data stays entirely on your machine.
Related Connectors
Mean Reversion Strategy MCP
Identify price extremes and mean-reversion entry points using Z-score and RSI filters.
Trend Filter Strategy MCP
Deterministic trading strategy using ADX for trend detection and Moving Averages for execution.
Document Metadata Flattener MCP
Flattens nested JSON metadata into dot-notation keys for vector store compatibility.
Variogram Modeling Analysis MCP
Compute experimental variograms and fit spatial continuity models.