# Honeywell Process MCP

> Honeywell Process connects your AI agent to deep industrial data streams. It lets you check asset health, create maintenance tickets, and analyze real-time production metrics for any Honeywell device. Instead of clicking through ten different dashboards, your agent pulls everything—from specific sensor readings to full shift reports—into one workflow.

## Overview
- **Category:** industry-titans
- **Price:** Free

## Description

This MCP gives your AI client direct access to the operational backbone of industrial facilities running Honeywell equipment. You can move past basic logging and actually manage complex workflows: checking if a machine needs immediate attention, pulling historical data on failures, or generating performance summaries for management. For instance, you don't have to manually pull records from a ticketing system and then cross-reference them with production output; your agent handles that whole sequence. This means your AI client can treat the entire facility—from the physical assets down to the network configuration—as one single data source. When connected via Vinkius, you get all these critical industrial tools in one place, letting your agent automate everything from scheduling preventive work to diagnosing an unexpected slowdown.

## Tools

### create_maintenance_ticket
Logs a new work order or ticket, assigning an issue to the appropriate maintenance team.

### get_asset_details
Retrieves in-depth information about any specific registered device, including its hardware specs and installation date.

### get_asset_health
Provides a real-time diagnostic score and status for an asset, identifying active faults or needed attention.

### get_maintenance_logs
Accesses full repair history, showing parts replaced, labor hours, and root cause analysis notes.

### get_operational_alerts
Lists current warnings or process deviations across the industrial system by severity level.

### get_process_metrics
Calculates and retrieves metrics like throughput rates, efficiency scores, and uptime percentages for production lines.

### get_production_data
Pulls raw output figures, comparing actual versus planned production targets by site or line.

### get_scan_events
Audits all barcode and RFID read events, showing timestamps and success/failure status for tracking purposes.

### get_shift_reports
Generates detailed summary reports covering output quantities, safety observations, and overall shift performance.

### list_assets
Lists all registered devices in the network, showing their type, location, and current operational status.

## Capabilities

### Diagnose immediate equipment issues
Quickly assess the operational status and identify fault codes for any registered asset.

### Build formal maintenance requests
Log new work orders, specifying assets, problem descriptions, and required completion dates.

### Track deep historical performance data
Retrieve detailed logs of past repairs, labor hours, and root cause analyses for compliance checks.

### Analyze operational throughput metrics
Get real-time and historical data on efficiency scores, uptime, and cycle times across production lines.

### Review full shift summaries
Generate comprehensive reports that combine output quantities, downtime events, and quality observations for a specific operational period.

## Use Cases

### Diagnosing a sudden drop in efficiency
The Operations Manager notices throughput dropped 20% yesterday. They ask their agent to analyze the issue. The agent uses `get_process_metrics` to find the dip, then runs `get_operational_alerts` to pinpoint if there was a connectivity failure or a component fault that caused it.

### Preparing for an annual compliance audit
The Maintenance Lead needs proof of service history. Instead of gathering physical records, they ask the agent to retrieve all relevant documents and summaries using `get_maintenance_logs` filtered by asset ID, ensuring every repair is documented.

### Investigating a high rate of scan failures
The Quality Analyst suspects scanning equipment. They tell their agent to pull data using the `get_scan_events` tool for the last week, quickly identifying which specific device ID is failing and why.

### Handing off shift responsibility
A departing crew member needs to leave a comprehensive handover. They ask the agent to generate a full summary by calling `get_shift_reports` combined with current asset status from `get_asset_health`, giving the next team all they need.

## Benefits

- Stop guessing if a machine is okay. By using `get_asset_health`, your agent instantly delivers an overall health score, telling you exactly what's wrong and whether it needs immediate attention.
- Never lose track of who fixed what or when. When you run `get_maintenance_logs`, you get the full audit trail—parts used, labor hours, even the root cause analysis—in one place.
- End manual reporting cycles. Instead of running separate reports for production and quality, your agent can pull data from `get_process_metrics` and combine it with `get_shift_reports` to give a full picture in seconds.
- When troubleshooting a slowdown, you don't check logs; you ask the agent. It runs checks like `get_operational_alerts` and cross-references them with raw data from `get_production_data` for immediate cause identification.
- Build an asset inventory on demand. The `list_assets` tool lets your AI client quickly map out every piece of equipment, showing its type and location without needing physical walk-throughs.

## How It Works

The bottom line is, your agent executes multi-step industrial processes that used to require jumping between separate databases, dashboards, and forms.

1. You tell your AI agent to perform an action, like 'Check the health of Scanner ID XYZ and create a ticket if it's bad.'
2. The agent uses the MCP to run multiple tools—like checking asset status first, then running diagnostics, and finally calling the ticketing tool.
3. The system returns a single, consolidated answer or confirmation, giving you an actionable report without needing manual data aggregation.

## Frequently Asked Questions

**How do I use get_asset_details with Honeywell Process?**
You provide a specific asset ID, and the tool returns comprehensive data on its hardware specs, firmware version, network configuration, and installation date. This is great for auditing.

**Can I check machine health using get_asset_health?**
Yes. You pass an asset identifier to `get_asset_health`, and it returns a real-time diagnostic score, component status, active fault codes, and predicted remaining useful life.

**What is the best way to track maintenance history with get_maintenance_logs?**
You need to specify the asset ID and ideally provide a date range. The tool returns all work order references, technician details, parts replaced, labor hours, and root cause analysis notes.

**How does list_assets help with inventory management?**
Running `list_assets` gives you an immediate roster of every registered device in the entire platform. It returns IDs, names, types, locations, and current status for a full overview.

**Do I need to use get_process_metrics for performance reports?**
Yes, `get_process_metrics` is how you evaluate operational performance. It calculates metrics like uptime percentages, efficiency scores, and cycle times across production lines.

**When should I use `create_maintenance_ticket` instead of just logging an issue?**
You initiate a formal, trackable work order with this tool. It requires the asset ID, problem description, and priority level. The system then automatically assigns a unique work order number and routes it to the proper maintenance team.

**How do I filter out non-critical alerts using `get_operational_alerts`?**
You specify filters like severity, status, or date range. This lets you narrow down incident history quickly, focusing only on 'Critical' warnings from the last 24 hours for example.

**Can I compare production output between different shifts using `get_production_data`?**
Yes, you can analyze and compare actual vs. planned production across multiple time periods. Specify the site, line, and date range parameters to see how performance stacks up against targets.