Safety Incident Report & Investigation: Capture, Investigate, Close

The problem this kills

Someone gets hurt on the floor, or a near-miss happens that should have been caught. What happens next decides whether you actually get safer — or just generate paperwork. Too often the report is a half-filled paper form that lands in a binder, the "investigation" is a hallway conversation that nobody wrote down, the corrective action is "talk to the operator," and nobody ever checks whether it got done. Three months later the same thing happens at the same machine.

Then the regulator, the insurer, or corporate asks for your log — and you're rebuilding the year from memory and a stack of forms. Was that laceration recordable or first-aid? Did you report the amputation within the required window? Who owned the guard-installation action, and did it actually close? When incidents are recorded inconsistently and corrective actions go untracked, you carry risk you can't even see, and your OSHA 300-style log is a fire drill instead of a byproduct.

You don't need a six-figure EHS platform to fix this. You can build the reporting-to-closure tool yourself, this weekend.

What you'll build

An internal web tool your team logs into. A supervisor or worker reports an incident in minutes — what happened, where, who was involved, the severity, a description, and photos straight from a phone into secure storage. Serious incidents fire an instant email alert to the EHS manager so nothing sits unseen over a weekend.

From there it follows a real workflow, not a filing cabinet. The EHS manager reviews each incident and classifies it — recordable, reportable, first-aid, or near-miss — using your rules, then approves that classification. The team runs a root-cause investigation (the why-behind-the-why) and writes corrective actions with an owner and a due date on each. EHS verifies the actions actually happened, then approves closure. At any point you export a clean incident register and an OSHA 300-style regulatory-log CSV — because the log was being built correctly the whole time, not reconstructed at the deadline.

What's inside the Implementation Plan

The downloadable plan is a single markdown file you paste into an AI coding agent. It opens by interviewing you about your own safety program — your locations and departments, your incident and injury types, exactly how you decide recordable vs first-aid, your reportable triggers and time windows, who classifies and who approves closure, and your messiest edge cases (contractors, multi-person incidents, late reports) — and then it reads a short spec back for your thumbs-up before it builds a thing. That's the difference between a tool shaped to your plant and a generic template you have to fight.

From there it walks the agent through the data model (incidents, people involved, photos, investigations, corrective actions, audit), the report form with photo upload, the serious-incident alert, the classification and approval gate, the investigation and corrective-action tracker, the verify-and-close gate, and the register and OSHA-style exports. Every step ends with a ready-to-copy prompt. There's a full "No API yet?" path: report incidents via the form or a CSV import, export a clean register CSV and regulatory-log CSV, and you never need an integration to ship.

The governance it includes (this is the point)

In EHS, the controls are the product. The plan builds them in: a login so only your team can use it; row-level security so each site or organization only ever sees its own incidents; a complete audit trail of who reported, classified, investigated, verified, and closed — and when; a hard human-in-the-loop approval gate so the EHS manager approves the classification and the corrective-action plan, and signs off on closure only after verification; and a duplicate guard on the incident number so the same event can't be logged twice. A near-miss can't be quietly "closed" without an investigation and a verified action. That's the audit story your regulator and your insurer want.

Who it's for

EHS managers, plant safety coordinators, supervisors, and safety-committee members who own incident reporting and corrective action and are tired of chasing forms and rebuilding the log at audit time. If you can explain to a new supervisor how an incident gets reported, classified, investigated, and closed, you can build this — no developer required.

You've got this — open the plan, paste the first prompt, answer a few questions about how your safety program actually runs, and you'll have your first incident reported, classified, and tracked to closure this weekend.