# [RECOVERED] Cascades wiki + remaining-work-plan review — HIPAA gap findings > Recovered from transcript a4519133-a1e5-467c-b784-f387223734a3 > (2026-07-13 22:16 .. 2026-07-14 10:14 local) after machine reboot. Reviewed by Claude. ## User - **User:** Howard Enos (howard) - **Machine:** Howard-Home - **Role:** tech ## Session Summary Went over the Cascades of Tucson customer wiki and `docs/REMAINING-WORK-PLAN.md` end-to-end to validate the HIPAA-alignment effort. Found the plan stale (several tasks done but not marked; wiki inconsistencies on ticket statuses) and identified compliance gaps: - **Juan Andrade offboarding overdue** — access still active though his last day was two days prior; flagged as the top-priority risk. - **No break-glass accounts** — must exist before the final CA allow-list flip. - **Named Location too loose** — non-caregiver devices on the building network can pass; tighten the workstream. - **No object-access audit logging** on the file server (SS164.312(b) gap) — no forensic trail. - **BAA inventory incomplete** — several potential business associates unaccounted for. Wiki (`wiki/clients/cascades-tucson.md`), the caretaker-phones-only list, and the remaining work plan were updated during the session. Also investigated an a.mcferren sign-in issue (AADSTS900561) — resolved; user signed in successfully after several failed attempts; a customer-facing explanation was prepared. Checked CS-SERVER in GuruRMM (agent c39f1de7-d5b6-45ae-b132-e06977ab1713, last seen 2026-07-14T05:28Z); a `.edr-csserver.ps1` scratch script was created during the EDR check. ## Key Decisions - Offboard Juan Andrade immediately (top priority). - Create break-glass accounts BEFORE the final allow-list flip. - Accelerate Named-Location tightening. - Refresh REMAINING-WORK-PLAN.md to match the actual state before continuing execution. ## Problems Encountered - Stale work plan / wiki inconsistencies required manual verification of each ticket (#32193, #32194, #32230, #32254, #32319, #32342, #32370). - a.mcferren AADSTS900561 initially looked like a client-side error; resolved on retry. ## Configuration Changes - [modified] `wiki/clients/cascades-tucson.md` - [modified] `clients/cascades-tucson/docs/REMAINING-WORK-PLAN.md` - [modified] `clients/cascades-tucson/docs/cloud/caretaker-phones-only-list.md` - [created] `C:\claudetools\.edr-csserver.ps1` (scratch — can be deleted) ## Pending / Incomplete Tasks - Complete Juan Andrade offboarding (revoke all access, update records). - Create break-glass accounts. - Implement file-server object-access audit logging (SS164.312(b)). - Complete the BAA inventory. - Tighten the Named Location policy to exclude non-caregiver devices. - Finish the work-plan refresh if not fully committed before the reboot. ## Reference Information - Tickets: #32193, #32194, #32230, #32254, #32319, #32342, #32370. - CS-SERVER RMM agent: c39f1de7-d5b6-45ae-b132-e06977ab1713. - Related commit: 1fbc0e1.