Security Onboarding New Client

Every new client arrives with an inherited security posture nobody has honestly assessed.
Run the structured intake across the four load-bearing controls — MFA, admin access,
backups, endpoint protection — and output a day-one risk list: what could hurt this client
this week, ranked, with owners. Work it in order:

1. Establish sources: what the previous provider or client handed over (documentation
   exports, admin credentials via secure handoff), what tooling is connectable now (RMM
   agents, Liongard where deployed), and what must be answered by the client's staff. Record
   the as-of date and source for every fact — inherited documentation is routinely stale, so
   verify rather than transcribe (read the live environment and identity posture in Liongard
   where available).
2. MFA coverage: percentage of user accounts with MFA enrolled and enforced (not just
   "capable"), the enforcement mechanism, and the exception list — service accounts,
   legacy-auth dependencies, "the owner doesn't like prompts" exemptions. Every exception is
   a risk-list candidate. Feed detailed follow-up to identity-mfa-health-check.
3. Admin inventory: every privileged account across identity platform, devices, firewall,
   backup console, and line-of-business admin panels. Flag: admin accounts used for daily
   work, shared admin credentials, ex-employee or previous-provider access still live, vendor
   accounts with standing access. Previous-provider access gets a dated transition plan. Feed
   to global-admin-audit for the deep pass.
4. Backup posture: what is backed up (and — the real question — what ISN'T: cloud mailboxes
   and file shares are commonly assumed-covered and aren't), where copies live (any
   offsite/immutable copy?), and the last successful restore TEST, not the last successful
   backup job. "Never tested" goes on the risk list at high rank.
5. EDR/endpoint protection presence: what product, on what percentage of the fleet (reconcile
   agent count against the RMM inventory — result-cap honesty on counts), who watches its
   alerts today, and unprotected classes (servers, Macs, that machine in the warehouse).
6. Quick environmental sweep for the classics: end-of-life OS in production, exposed
   remote-access (RDP/VPN with no MFA), credential spreadsheets (flag location, never
   contents), and email-security basics (route the SPF/DMARC read to dmarc-spf-failure-triage
   territory).
7. Produce the day-one risk list — the deliverable: each entry as risk / evidence / exposure
   in plain terms / recommended action / suggested owner, ranked by exploitability-now rather
   than audit-category. Cap it at the ten that matter. Open remediation tickets for the top
   items the client approves, and hand the list to the account manager for the roadmap
   conversation (cyber-risk-posture-review for the ongoing cadence).

Guardrails — always:
- Verify, don't transcribe: inherited docs and previous-provider claims get checked against
  live reads or client confirmation; every finding carries its source and as-of date.
- Never present partial visibility as a clean bill — "no EDR gaps found among the 60% of
  devices reporting" is the honest sentence.
- Findings are stated neutrally (this is an intake, not an indictment of the previous
  provider) and factually per the defensive-writing-standard; the client may share this
  document.
- Credentials received in handoff go straight into the desk's credential store — never into
  tickets or intake notes; credential spreadsheets found are flagged by location only.
- The risk list recommends; the client decides — remediation beyond the agreed onboarding
  scope is quoted work, flagged to the account manager, never silently absorbed or skipped.
- Degradation: without RMM/Liongard connectivity, the intake runs on exports and client
  interviews — mark every unverified item as unverified, and re-run the checks once tooling
  lands.