- CLAUDE.md: triggers now query coordination API (/api/coord/status,
/api/coord/components, /api/coord/messages) instead of reading
PROJECT_STATE.md files
- COORDINATION_PROTOCOL.md: new doc covering locks, component states,
workflows, work items, and inter-session messages via ClaudeTools API
- guru-rmm/PROJECT_STATE.md: marked ARCHIVED, redirects to
COORDINATION_PROTOCOL.md for live state
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds /api/coord/* endpoints for real-time cross-session coordination:
- coord_workflows: named units of work per project
- coord_work_items: tasks within workflows with dependency chains
- coord_session_locks: exclusive resource locks with auto-expiry (TTL)
- coord_component_states: live component state per project (upsert)
- coord_messages: cross-session messaging and broadcasts
- /api/coord/status: cross-project snapshot endpoint
Replaces PROJECT_STATE.md as the coordination layer for Claude sessions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Deployed radio-show FastAPI redesign (HEAD already at b008b61 with sort fix) to Jupiter; rebuilt radio-archive container.
- Solved Jupiter audio 404 by rsync IX -> Jupiter over LAN (8.09 GB, ~75s @ 108 MB/s); installed Jupiter root pubkey on IX root for passwordless server-to-server access.
- Addressed 6 Note-for-Mike blocks from Howard (Cascades SDM activation root cause, IMC1 AIM SQL diagnosis correction, Sombra/Transwiz patterns, Stamback prepay).
- Restored dead Discord bot (silent since 2026-05-06 reboot); installed as NSSM service ClaudeToolsDiscordBot with auto-restart + log rotation.
- Resolved /sync conflict on memory entry by dropping redundant local commit in favor of Howard's richer feedback_syncro_appointment_owner.md.
- Kicked off Apple Developer Program enrollment (HH5UA87LAH); flagged D&B name mismatch (DUNS 005661506 registered to 'COMPUTER GURU' not 'Arizona Computer Guru LLC') as real blocker; vaulted full sequence at infrastructure/apple-developer-program.sops.yaml in vault repo.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Add hard rule: 9269129 (Prepaid Project Labor) is Exempt and does NOT deduct
from prepay_hours block — never use for normal work (verified 2026-05-04)
- Expand prepay_hours check from emergency-only to ALL billing workflows
- Fix emergency/prepaid branching table to use delivery-channel product instead
of hardcoding 26118 (Onsite) for remote and other labor types
- Clarify invoice step 15: $0.00 invoice total is correct for prepaid customers;
verify by checking customer.prepay_hours dropped by quantity
- Field 7 (Assigned Tech): add explicit default to API key owner; mark as MUST
always be included in POST payload to prevent null user_id on ticket create
- Add billing workflow hard rule: read prepay_hours before any billing, not just
emergency, so prepaid invoice behavior is known before execution begins
Triggered by ticket #32265 (Russo Law Firm) missing assignee/priority/billing.
Russo Law has 12.5 prepaid hrs — 0.5 hrs correctly deducted via invoice #67578.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Exchange REST API still propagating (28 min). Need manual verification via
Exchange Admin Center to unblock HIPAA compliance check.
Instructions provided:
- Access Exchange Admin Center
- Search for Britney Thompson mailbox
- Document litigation hold status (enabled/disabled, date, duration)
- Report findings back in repo
Priority: HIGH - blocks Wave 1 caregiver rollout planning.
HIPAA requirement: §164.308(a)(3)(ii)(C) + §164.316(b)(2)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
9-hour day on Cascades caregiver phone Shared Device Mode activation.
Root cause of repeated AADSTS50097 was missing Cloud Device Administrator
role -- pilot.test cannot self-register devices for shared mode. Created
dedicated devices@cascadestucson.com (CDA role, MFA on Howard's phone).
Final attempt on Phone A produced an Entra device record with shared-mode
markers (registeredOwners=0, registeredUsers=0). Resume tomorrow by
signing pilot.test in to verify SDM is actually active.
Side wins: ALIS SSO Entra App Registration created (vault commit 90ada33,
blocked on Medtelligent enabling App Store side); 2 of 3 caregiver CA
policies flipped from Report-only to Enforced; kiosk profile bumped to
v13 with full Android nav bar, 12hr inactivity signout, 6-app allowlist
including Company Portal.
Microsoft ticket #2605070040009774 still open.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
New client onboarding for The Law Offices of Chris Scileppi with initial
session log documenting diagnosis on Sylvia's Mac mini (Mac14,3, M2, 8 GB).
Issue: System running out of memory; Apple Mail footprint thrashing the box.
Two Envelope Index rebuild attempts confirmed the mailbox itself exceeds what
8 GB can hold. Disabled Mail at the OS level, moved user to webmail, and
recommended replacement with an M4 Mac mini (16 or 24 GB).
Ticket #32262 resolved. 1 hr onsite logged but deliberately not invoiced.
Files:
- clients/scileppi-law/PROJECT_STATE.md
- clients/scileppi-law/docs/overview.md
- clients/scileppi-law/docs/issues/log.md
- clients/scileppi-law/session-logs/2026-05-07-howard-sylvia-mac-mini-mail-memory.md
All 5 ComputerGuru apps successfully onboarded:
- Security Investigator, Exchange Operator, User Manager, Tenant Admin, Defender Add-on
- API permissions granted (0 errors)
- Exchange Administrator role assigned to Security Investigator SP
Exchange REST API access pending propagation (15-30 min typical).
Next: Re-test Exchange REST after 09:30 AM MST to verify litigation hold check.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Cannot verify litigation hold status - ComputerGuru Security Investigator
app not onboarded to Cascades tenant (HTTP 401 on Exchange REST).
User account confirmed (Britney.Thompson@cascadestucson.com).
Next steps:
- Onboard Security Investigator app to tenant
- Assign Exchange Administrator role
- Re-run litigation hold verification
HIPAA compliance blocker per Howard's 2026-05-06 note.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Approved:
- Memory caps: SQLEXPRESS 12GB, WID 512MB, AIMSQL 256MB
- AIMSQL consolidation (pending backup)
- AD is in use, WSUS is not
Howard may proceed with implementation.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Lauren Hasselman could not create a Teams group on 2026-05-05.
Diagnostic confirmed the block is at the Teams Admin policy layer
(intentional, gated on HIPAA prerequisites in m365.md issues #12-#14),
not an Entra/M365-Group permissions defect. New teams-rollout.md
captures prerequisites, HIPAA config checklist, canary test plan
(Lauren as primary canary), and exit criteria. Linked from m365.md
issue #14.
Follow-up on three pending items from breach check:
- IdentityRiskyUser scope: consented but requires P2 license
- Dime Client app: internal app requiring verification with Dan Center
- Microsoft Authenticator: drafted upgrade plan and recommendations
Created comprehensive follow-up report with action items.
Machine: Mikes-MacBook-Air
User: Mike Swanson (mike)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Darkweb scan follow-up: ran 10-point breach check on jantar@dataforth.com (no IOCs),
revoked eM Client OAuth grant and app role assignment, disabled eM Client SP tenant-wide.
Syncro ticket #109790034 created, billed 1hr prepaid, resolved.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Mike's 4/30 audit (surfaced via /sync) flagged that 31 closed tickets had
00:00:00 in Syncro time tracking — bare add_line_item bypasses time entries
and breaks reporting. I had just done the same on today's 3 tickets; Winter
retroactively added time entries. Rewrote the syncro skill (commit ec98c6c)
to make timer_entry -> charge_timer_entry the default and demote bare
add_line_item to a fallback for non-time items only. Disabled the
now-redundant scheduled agent (trig_01CAfvwoQ4nLcKEqbU4UQmSa).
Deletions (~1,500 lines of dead docs):
- .claude/hooks/ — docs-only directory, no executables. Referenced scripts
setup-context-recall.sh / test-context-recall.sh did not exist. Hooks
would have POSTed to localhost:8000; the API actually ran at
172.16.3.30:8001 and is no longer in use.
- .claude/AUTO_CONTEXT_SYSTEM.md — 347-line duplicate spec of CLAUDE.md's
Automatic Context Loading section, referencing unimplemented hooks.
- .claude/URGENT-vault-path-bug.md — 217-line urgency note for a fix that
already shipped weeks ago.
- .claude/context-recall-config.env.example — config template for the same
dead system.
Refactors (~500 lines net removed):
- /save and /sync now wrap bash .claude/scripts/sync.sh as the single
source of truth for git ops. /save adds a session-log-writing step in
front; /sync invokes the script directly.
- Dropped /sync's manual git phases that contradicted sync.sh.
- Dropped the cp -r ~/ClaudeTools/.claude/commands/* ~/.claude/commands/
step (clobbered per-user customization in the multi-user model).
- Dropped auto-invoke of /refresh-directives (command does not exist).
- Dropped references to directives.md (file does not exist).
- /save now documents the rm -f save_narrative_prompt.txt step, fixing
the stale-prompt bug Howard documented in feedback_tmp_path_windows.md.
Fixes:
- CLAUDE.md SESSION_STATE.md reference replaced with the canonical
PROJECT_STATE.md (per-project, with protocol at
.claude/PROJECT_STATE_PROTOCOL.md). 16 client folders already use
PROJECT_STATE.md; SESSION_STATE.md was only a stale reference.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
