MCP For Claude Marketplace
Outlier Detection Engine

Outlier Detection Engine MCP Connector for Claude

A+

Identify statistical anomalies in massive datasets local using deterministic Z-Score and IQR methods. Stop LLMs from guessing which rows are outliers.

1 tools Official Updated Oct 1, 2026 Official Vinkius Partner
More Details Connect to Claude

Outliers skew machine learning models and corrupt statistical analysis. If you ask an LLM to scan 10,000 rows for anomalies, it will exhaust its context and arbitrarily flag random rows based on visual intuition — not math.

This MCP delegates outlier detection to simple-statistics. The engine calculates exact Means, Standard Deviations, and Quartiles, then flags specific rows mathematically using Z-Score or IQR bounds. No intuition, no guessing — just pure deterministic statistics.

The Superpowers

  • Mathematical Precision: Every flagged outlier comes with its exact Z-Score or IQR boundary values.
  • Multiple Methods: Choose Z-Score (parametric, best for normal distributions) or IQR (robust, best for skewed data).
  • Customizable Threshold: Set your own sensitivity (Z > 3, IQR × 1.5, etc.).
  • High Performance: Scans thousands of rows instantly on your local machine.
statistical-analysisanomaly-detectionz-scoreiqrdata-cleaningmath-engine

1 tools expose this connector's capabilities to your AI agent.

detect_outliers

Deterministically identify statistical outliers in datasets using Z-Score or IQR methods

See how to talk to your AI agent using Outlier Detection Engine.

Find all rows where the 'Temperature' reading is a statistical outlier using Z-Score > 3.

Found 4 outliers. The most extreme is row 142 with Temperature 98.5°C (Z-Score = 4.1), followed by row 87 (Z = 3.8), row 201 (Z = 3.4), and row 15 (Z = 3.1).

Check the 'Price' column for anomalies using the robust IQR method with a 1.5 multiplier.

Using an IQR threshold of 1.5, I identified 12 items priced significantly above the upper bound of $450 (Q3 + 1.5×IQR). These appear to be luxury or premium-tier products.

Are there any abnormal network latency values in this monitoring dataset?

Yes. Using Z-Score analysis, 3 network requests had ping times exceeding 3 standard deviations (Z > 3): rows 44 (Z=3.9), 128 (Z=3.5), and 302 (Z=3.2).

Z-Score assumes data is normally distributed and is sensitive to extreme outliers. IQR is based on percentiles (25th and 75th), making it robust and ideal for skewed or non-normal data.

Related Connectors

Observability & Tracing Calculator

Observability & Tracing Calculator MCP

3 tools Official

Deterministic engine for distributed tracing metrics, cost estimation, and anomaly detection.

A+ View details →
Agent Self-Reflection Sentiment Scanner

Agent Self-Reflection Sentiment Scanner MCP

3 tools Official

Analyzes agent execution logs to detect self-correction markers and measure stability.

A+ View details →
History Matching Quality

History Matching Quality MCP

4 tools Official

Quantifies the alignment between simulated reservoir production and observed historical data.

A+ View details →
Sliding Window Aggregator

Sliding Window Aggregator MCP

3 tools Official

Deterministic engine for sliding window metric aggregation, trend analysis, and anomaly detection.

A+ View details →