We won’t publish a security overview until we can back it with evidence.
Most of MediFlow’s security posture lives in code today — organization isolation, server-side role checks, PHI-safe audit metadata, AI minimization gates. The formal overview will be published when each control can point to tests, screenshots, and provider agreements rather than claims.
Implemented in code. Verified by request.
Live status of MediFlow’s security controls. Implemented means the control is enforced in code. Verified will appear here when third-party evidence is on file.
Organization isolation
Every subscription maps to an organization workspace. PHI and operational data are scoped to that workspace.
Role-based access
Owner, Admin, Clinician, Staff, Billing, Readonly. Checks enforced at the server/function boundary.
Audit foundations
Clinical, billing, membership, AI, and admin events modeled with actor context and PHI-safe metadata.
PHI handling
Server-side mutations only. Fail-closed audit writes block the mutation when audit fails.
Authentication & sessions
Unique user identities. MFA enforcement is treated as a launch gate, not a post-launch ask.
AI privacy controls
Server-side execution, feature-flagged, quota-aware, PHI-minimized, human-reviewable outputs.
Document security
Time-bound document URLs, server-side access checks on upload and read.
Monitoring & incident response
Runbooks, on-call rotation, retention schedule, breach notification timelines.
External evidence
BAA, DPA, subprocessor inventory, penetration testing, SOC 2 readiness, network allowlists.
This page will not assert completed security certifications, full production hardening, penetration testing, end-to-end encrypted PHI, external audit streaming, network allowlists, or enterprise identity provisioning until those controls are implemented, verified, and contractually supported.
If you need a current readiness statement before launch, ask — we’ll share what’s real today and what is still being built.
Ask for the current readiness statement