predictive-maintenance-mcp

Overview

Predictive Maintenance MCP Server is an open-source framework that integrates Large Language Models with predictive maintenance and fault diagnosis workflows. It provides 52 specialized MCP endpoints (46 tools, 2 resources, 4 prompts) for signal acquisition, spectral analysis, diagnostics, reporting, prognostics, and decision support. The server enables engineers to describe their needs in plain language, and the AI calls appropriate analysis tools to deliver professional diagnostic results including bearing fault detection, risk assessment, anomaly detection, and remaining useful life estimation.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• load_signal

  Load vibration signal file (CSV, WAV, MAT, NPY, Parquet)

  Try:Load real_train/OuterRaceFault_1.csvTry:Load the bearing signal from data/test_signal.csv
• analyze_fft

  Frequency spectrum analysis with automatic peak detection

  Try:Analyze FFT of the loaded signalTry:Check frequency spectrum for peaks
• analyze_envelope

  Envelope analysis for bearing fault detection

  Try:Perform envelope analysis on the bearing signalTry:Check for bearing faults using envelope
• diagnose_vibration_tool

  Integrated evidence-based vibration diagnosis pipeline

  Try:Diagnose the bearing for faultsTry:What's wrong with this vibration signal?
• calculate_bearing_characteristic_frequencies

  Compute expected fault frequencies from bearing geometry

  Try:Calculate bearing fault frequencies for 6205-2RS bearingTry:What are the characteristic frequencies for this bearing?
• check_bearing_faults_direct

  Multi-fault detection (inner/outer/ball/cage) in bearings

  Try:Check for bearing faults in the signalTry:Detect which bearing fault is present
• generate_diagnostic_report_docx

  Generate structured Word document diagnostic report

  Try:Create a Word report of the diagnosisTry:Generate professional diagnostic document
• train_anomaly_model

  Train novelty detection model on healthy baseline signals

  Try:Train anomaly detector on healthy signalsTry:Create ML model for detecting anomalies
• estimate_rul

  Estimate Remaining Useful Life using various models

  Try:Estimate remaining useful life of the bearingTry:How long will this component last?
• generate_maintenance_recommendations

  Generate context-aware maintenance recommendations from diagnosis

  Try:What maintenance actions should I take?Try:Recommend maintenance based on diagnosis
• search_bearing_catalog

  Search bearing specifications by model number

  Try:Find specs for 6205-2RS bearingTry:Search for bearing model 6303
• extract_features_from_signal

  Extract 17+ statistical and spectral features from signal

  Try:Extract features from the vibration signalTry:Get all relevant features from the data

Comparable tools

Installation

pip install predictive-maintenance-mcp

{
  "mcpServers": {
    "predictive-maintenance": {
      "command": "/full/path/to/uvx",
      "args": ["predictive-maintenance-mcp"],
      "env": { "UV_LINK_MODE": "copy" }
    }
  }
}

git clone https://github.com/LGDiMaggio/predictive-maintenance-mcp.git
cd predictive-maintenance-mcp
.\setup_claude.ps1

FAQ

What file formats does the MCP server support for vibration signals?

The server supports CSV, WAV, MAT, NPY, and Parquet formats for vibration signal files.

Can this MCP server be used with other LLM assistants besides Claude?

Yes, the predictive maintenance MCP server can be used with any LLM that supports the Model Context Protocol, including local setups like Ollama.

On Hacker News

Recent discussion from the developer community.

• Show HN: MCP server for vibration-based predictive maintenance▲ 2💬 3
  Story by lgdimaggio · 2025-12-27

Compare predictive-maintenance-mcp with