Incident Severity Triage & Routing

The problem this kills

When an incident or near-miss comes in, the clock is already running - but the severity call usually isn't consistent. One supervisor logs a slip as "minor," another logs the same thing as "serious." The right owner finds out late, the notify list is whoever someone happened to remember, and there's no clean record of who decided what or when. By the time it's escalated, the SLA is already blown.

The fix isn't another inbox or a heavier spreadsheet. It's a small tool that proposes a severity from your matrix the same way every time, suggests the right owner and notify list from your routing rules, and then puts a human in charge of the final call before anything goes official - with the escalation clock visible the whole time.

What you'll build

A private, login-only web app for your EHS leads, duty managers, and operations team:

• Intake - incidents arrive from a simple form or a CSV import (one row per incident), deduplicated by your incident ID.
• Proposed severity - the tool reads your configurable matrix and proposes both an actual and a potential severity, with the reasoning shown.
• Suggested routing - based on severity and area, it suggests the correct owner, notify list, and SLA from your routing table.
• The human gate - the duty manager confirms or overrides the proposed severity, with a logged reason, before anything is sent.
• Alerts with a clock - on confirmation, Resend notifies the owner and list, and an acknowledgment SLA timer starts.
• Escalation - if the incident isn't acknowledged in time, it escalates to the next tier automatically.
• Export - a clean triage-log CSV of every decision, override, and timestamp for your records or your system of record.

What's inside the Implementation Plan

• It starts by interviewing you about your business. Before writing a line of code, the plan has the AI agent ask about your current intake process, your real severity matrix, your area and code conventions, your routing rules, your SLA tiers, and your messy edge cases - then it reflects a short tailored spec back to you for a thumbs-up. You get a tool shaped around how you actually triage, not a generic template.
• A step-by-step build, each step ending in a ready-to-paste prompt for your AI coding agent.
• A configurable severity matrix and routing table you can edit without touching code.
• The duty-manager confirm/override gate with mandatory logged reasons.
• Resend alerts, SLA timers, and tiered escalation on missed acknowledgment.
• A "No API yet?" fallback so you can build and run the whole thing today from a CSV, and export back to CSV.
• A verification checklist so you know it actually works before you trust it.

The governance it includes (this is the point)

• Login so only your team can open the tool.
• Row-level security so people only ever see their own organization's incidents.
• A complete audit trail - every proposal, confirm, override (with reason), notification, and escalation is logged with who and when.
• A hard human-in-the-loop gate - the AI proposes a severity and routing, but nothing becomes official until the duty manager approves; only then do alerts go out.
• Duplicate guards - the same incident ID can't be triaged or routed twice.

Who it's for

EHS leads and operations managers who need consistent severity calls and fast, rule-based escalation - without waiting on IT or buying a heavyweight EHS platform. If you can describe your matrix and your routing rules, you can build this.

You've got this - paste the first prompt and let the interview tailor the rest.