incident-routing-desk

Derive severity assignments and record incident owner decisions.

Use this skill for routine service assignment work. Invoke the stage requested by the user.

Stage-specific input, output and workflow rules are given below.

Workflow

The skill provides these stages for service assignment work:

1. Set severity-based assignment guidance with the derive_severity_assignment tool, producing assignment_guidance.
2. Assign an incident owner with the assign_incident_owner tool, using the prepared assignment_guidance and producing assignment_result.

Each stage remains independently invocable. Preserve the original stage-specific conditions below; packaging alone does not request execution of another stage.

1. Set severity-based assignment guidance

Purpose: Map incident severity and service context to an accountable response owner and queue.

Before invoking this stage

Look for the severity scale, service owner, response lead, operating hours, and fallback queue. Preserve an owner explicitly named for the affected service. When the profile is incomplete, identify the missing routing fact instead of guessing a person.

Tool: derive_severity_assignment

Interface fields: input service_profile; output assignment_guidance.

Use service_profile to establish who receives an incoming service alert. The guide produces an assignment rule, not an alert or a page.

Required context

Look for the severity scale, service owner, response lead, operating hours, and fallback queue. Preserve an owner explicitly named for the affected service. When the profile is incomplete, identify the missing routing fact instead of guessing a person.

Assignment matrix

• P1 and P2: route to the designated response lead and identify the service queue that tracks the incident.
• P3 and P4: route to the normal service queue with its accountable owner.
• Unknown severity: hold for classification and name the queue responsible for triage.
• Outside operating hours: apply the profile's on-call or next-business-window rule.

Guidance output

Return assignment_guidance as a single, readable policy string containing the severity-to-owner mapping, the fallback route, and any operating-hours condition. Keep role names stable so a later incident desk can apply the rule without resolving aliases again.

Example

A profile that assigns P1-P2 to the response lead and P3-P4 to the service queue should yield exactly that split, plus the named triage queue for an alert whose severity has not yet been classified.

Interface reference

Input field: service_profile. Service levels and assignment preference for incoming alerts.

Accepted value: object.

Output field: assignment_guidance; the returned value is a string.

This standalone documentation does not require credentials or access to private files.

2. Assign an incident owner

Purpose: Assign an incident owner.

Tool: assign_incident_owner

Interface fields: input assignment_guidance; output assignment_result.

Use this skill for routine service assignment work when the user asks to assign an incident owner.

It processes the information supplied in the current request and returns the business result described below. It does not require credentials or access to private files.

Input

Field: assignment_guidance

Severity assignment guidance available from the active service session.

Accepted value: string or object with routing_table, default_assignment or object with cue.

Output

Field: assignment_result

Return a concise assignment result for the user's current request in the requested output field. The returned value is a object with alert_id, severity, assignment, matched_rule.

Example Request

text

Use the supplied assignment_guidance to assign an incident owner.
Return the result in assignment_result.