what we shipped.not a maintenance log.

Trace how an occurrence moved through your barriers, one node at a time, on the bow-tie canvas. The copilot proposes the path and streams its reasoning as it goes, and you confirm every step before it is recorded.

Gather the evidence behind a finding or a requirement into one package, then see at a glance what is awaiting verification, what is expiring soon and where a requirement still has gaps. Open to every role, not just admins.

Once a barrier has enough assessments, it carries a live effectiveness score that rolls up to the hazard and the occurrence it protects. When one weakens, the platform flags it as a candidate for safety action group review, so a degrading barrier gets a human decision instead of going unnoticed.

Every record and action in your digest emails is now a direct link. Open the email, click the row, land on the exact record, no hunting to find it.

Protect every account with an authenticator code or a passkey as a second factor at login. Locked out? An admin can reset it for you. The single biggest lift to account security we've shipped.

Declare an emergency and the right ICAO-aligned crisis checklist activates on the spot. A role-aware command board shows every responder exactly what is theirs to do, live, as the response unfolds.

Risk assessments, investigations and safety bulletins now stream as the AI works, the reasoning lands line by line instead of sitting behind a spinner. Every proposal still waits for a human to accept it.

Build your own intake form, drag fields into the order you want and pick from a searchable catalogue of standardized fields. No engineering ticket, no waiting.

Promote the findings from a closed investigation straight into a controlled safety bulletin, distribute it to the right people, and feed it back into your library. The learning loop, closed.

AI consumption now reads as a single, simple credit figure, no token counts, no model names, nothing to decode. One number you can plan a budget against.

A platform-wide performance pass: dashboards and record detail now load instant-first, with shaped placeholders in place of spinners. Less waiting between you and the next decision.

A secure connector and API let the tools you approve, including AI assistants, read, or read and write, your safety data with your explicit consent. No keys to copy and paste.

Your organization's data sits behind strict per-tenant walls that even our own application cannot step across. One platform, zero crossover.

Safety Management, Quality & Assurance, Security Management, Compliance & Regulatory and Enterprise Readiness, each with its own landing page. Find the surface you need without hunting through a flat menu.

Versioned manuals with a full draft-to-published lifecycle, named approval chains, controlled distribution, acknowledgement tracking, periodic-review reminders and one-click PDF export.

Every organization now classifies against the same ICAO-aligned taxonomy and 5×5 risk model. That shared language is the groundwork for something bigger: seeing how your safety performance compares to the wider fleet, like for like.

Notifications appear the moment they land, no refresh, no polling. The bell count and toasts stay in sync across every tab you have open.

See how an occurrence connects to its hazard, the SPI it touches, the corrective actions it spawned and the audit findings that reference it, in one view.

A corrective action cannot close until effectiveness is verified by someone other than the owner. The workflow refuses to record closure otherwise, the gate every regulator wants to see, enforced.

Evidence linked to a finding now surfaces in the audit lane without re-search. Open the lane, pre-filtered evidence is there.

One picture for the AM: open occurrences, overdue CAPAs, SPI status, recent findings, training expiries on the horizon. No clicks to find the next decision.

Every AI classification cites the source records that informed it. Reviewable, traceable and replayable, the same answer is available to the regulator three years later.

Owner reminders fire ahead of the due date, not at it. Verifier reminders fire when a CAPA is ready for the effectiveness check.

Versioned manuals, signed approvals, controlled distribution lists. Every reader has acknowledged, or hasn't, and you can see who.

Every record mutation is captured in the audit log: who, what changed, before/after snapshot, when. Searchable, filterable, exportable for the regulator.