• TYPES
  • LEVELS
  • FIELDS
  • HUMAN

How to audit incident configurations with your AI

Connect Incident.io to your agent to pull your current setup data. Once connected, you can run audits on your configuration without clicking through menus.

Ask AI about this page

Short answer

How can I audit my incident setup using AI?

Your agent pulls your incident types, severity levels, and custom fields directly from Incident.io. You'll get a clear view of your current configuration to find gaps or inconsistencies. Here is how the process works.

Audit outcomes.

Where your configuration data lands.

The agent organizes your setup data into these specific views.

TYPES

Incident type list

The agent pulls every incident type you have defined. You can see if your categories match your current operational needs.

LEVELS

Severity mapping

The agent lists all available severity levels. This helps you confirm that your escalation paths align with your priority definitions.

FIELDS

Custom field inventory

The agent identifies all custom fields currently in use. You can check if required data points are missing from your forms.

HUMAN

Manual adjustments

The agent only reads your configuration. You must still go into the Incident.io UI to change or add new fields and types.

The workflow

What your AI does when the audit starts.

The agent performs a series of read-only checks to build a map of your environment.

  1. Identify incident types

    The agent fetches the full list of incident categories you use to classify issues.

    list_incident_types

  2. Check severity levels

    The agent retrieves the specific severity tiers defined in your workspace.

    list_severities

  3. Inventory custom fields

    The agent scans for all custom fields used to collect extra data during an incident.

    list_custom_fields

  4. Analyze for gaps

    The agent compares the gathered data against your specific audit questions to find missing fields or mismatched types.

    list_incident_typeslist_severitieslist_custom_fields

Try it

Copy these to start.

Use these prompts to begin your audit.

Starting points

These are basic starting points. You can add specific names of incident types or fields to make the questions more targeted.

Incident.io Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_SYXytajPn2E8oSQTe6uxrJcvhwsU2x1pHkOPCe4y/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Incident.io capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "incidentio-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_SYXytajPn2E8oSQTe6uxrJcvhwsU2x1pHkOPCe4y/mcp"
    }
  }
}
Copy into chat04

  • List all my incident types and severities in Incident.io.

  • What custom fields are currently configured in my Incident.io setup?

  • Compare my incident types against my custom fields to see if any types are missing data points.

  • Do I have a severity level for 'Critical' defined in my configuration?

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Incident.io once, then ask.

Just follow the link to connect your account. Your credentials stay encrypted, and you can immediately start asking questions about your setup.

Connect Incident.io to your AI

FAQ

How this task behaves.

  • 01

    Can the AI change my incident types?

    No. The agent can only read your current configuration. You have to make any changes manually in Incident.io.

  • 02

    Can the AI create new custom fields?

    No. The available tools only allow the agent to list existing fields, not create or modify them.

  • 03

    Does the AI see my actual incident data?

    This specific audit task only looks at your configuration, not the individual incidents themselves.

  • 04

    Can I use this to find missing fields?

    Yes. By listing your custom fields and incident types, the agent can help you identify where your data collection might be incomplete.

  • 05

    Will this work in Claude or Cursor?

    Yes, it works in any MCP-compatible client like Claude, Cursor, or Windsurf once you have connected the Connector.

  • More questions about Incident.io? The Connector page answers them. See everything the Incident.io Connector can do

Connect Incident.io to Claude, Cursor, ChatGPT & more