agent: suppress distress dispatch when the auth failure came from a dead endpoint

Follow-up to 13dd6723: a platform-403 (stopped/retired app, HTML error
page) now skips the DeviceNotRegistered/AuthCertificateRejected distress
report - it would be sent to the very endpoint that just answered with a
platform page, and its classification would be wrong. Counting and the
shutdown thresholds are unchanged, so a dead endpoint still drives the
agent to its soft shutdown instead of retrying forever.

The one-report-per-streak gate moves from "count == 1" to a streak-scoped
flag: an endpoint-unavailable failure no longer consumes the streak's
report slot, so a later genuine 403 from a live backend still dispatches.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

.github/workflows/build-agent.yml

@@ -270,6 +270,12 @@ jobs:
             if (Test-Path "scripts/Bootstrap/Test-ShouldBootstrapAgent.ps1") {
               Send-Blob -Local "scripts/Bootstrap/Test-ShouldBootstrapAgent.ps1" -BlobName "Test-ShouldBootstrapAgent.ps1" -ContentType "text/plain; charset=utf-8"
             }
+
+            # Customer-side Graph add-on grant script -- docs and admin UI link
+            # https://download.autopilotmonitor.com/agent/Grant-AutopilotMonitorAddOn.ps1
+            if (Test-Path "scripts/CustomerSetup/Grant-AutopilotMonitorAddOn.ps1") {
+              Send-Blob -Local "scripts/CustomerSetup/Grant-AutopilotMonitorAddOn.ps1" -BlobName "Grant-AutopilotMonitorAddOn.ps1" -ContentType "text/plain; charset=utf-8"
+            }
           }
 
           # Update AdminConfiguration LatestAgentV2* -- ONLY on stable cutover. The hash-oracle

.github/workflows/ci.yml

@@ -107,8 +107,22 @@ jobs:
       - name: Typecheck
         run: npx tsc --noEmit
 
+      # The vector-search tests tokenize with the real all-MiniLM-L6-v2 model
+      # (~23 MB tokenizer + ONNX weights). transformers.js caches under
+      # node_modules by default, which npm ci recreates — so every CI run
+      # re-downloaded from huggingface.co and rate-limits (429) failed the job
+      # (run 31252869254). Persist the model across runs instead; HF_CACHE_DIR
+      # is honored by vector-search-provider.ts.
+      - name: Cache embedding model
+        uses: actions/cache@v4
+        with:
+          path: ${{ github.workspace }}/.hf-cache
+          key: hf-model-Xenova-all-MiniLM-L6-v2
+
       - name: Tests
         run: npm test
+        env:
+          HF_CACHE_DIR: ${{ github.workspace }}/.hf-cache
 
   guardrails-in-sync:
     name: Generated guardrails match rules/guardrails.json

docs/agent/continue-anyway-observation.md

@@ -0,0 +1,114 @@
+---
+type: Concept
+title: Continue-Anyway Observation Mode — Device-Phase ESP Failure Defang
+description: Operator-set tenant opt-in that turns an immediate Device-phase ESP terminal failure on a Continue-Anyway-enabled profile into a bounded 60-minute observation; the DAD-validated real-user desktop plus the Hello gate detect the true end, and the session completes as Succeeded with an esp-soft-failure marker.
+resource: /src/Shared/AutopilotMonitor.DecisionCore/Engine/DecisionEngine.Edge.cs
+tags:
+  - agent
+  - decision-engine
+  - esp
+  - continue-anyway
+  - soft-failure
+  - tenant-config
+timestamp: 2026-08-08T00:00:00+02:00
+---
+
+# Continue-Anyway Observation Mode
+
+## Problem
+
+A Classic user-driven enrollment whose Device ESP hits the configured terminal timeout
+(`syncFailureTimeoutMinutes`, typically 30) fails **before AccountSetup is ever entered**.
+The classic advisory defang (`HandleEspTerminalFailureV1`) requires
+`AccountSetupEnteredUtc != null`, so these sessions hard-fail immediately — even when the
+ESP profile allows "Continue anyway" and users demonstrably dismiss the failure screen and
+reach a working desktop. A fleet held at the timeout wall by one slow blocking app then
+shows ~100% Failed (tenant c9787ba2: 948/1000 `esp_terminal_failure`, culprit "Encompass
+Hybrid Installer"), which tells the tenant nothing.
+
+## Mechanism
+
+Opt-in flow (default off — absence keeps the immediate hard fail):
+
+1. **Tenant setting** `EnableEspContinueAnywayObservation` (`TenantConfiguration`,
+   operator-set only, NOT exposed in the tenant admin settings UI; the toggle lives in
+   the Global-Admin tenant management modal next to Bootstrap Token / Unrestricted Mode,
+   and the field is GA-gated on both write paths — the generic PUT silently reverts it
+   for tenant admins, the field patch rejects it with 400) → served via `agent/config`
+   (`AgentConfigResponse`, ConfigVersion 37).
+2. **Agent** stamps `espContinueAnywayObservationEnabled=true` into the
+   `EspConfigDetected` signal payload (both emitters: the orchestrator FirstSync
+   bootstrap and `DeviceInfoCollector.PostEspConfigDetectedSignal`) — only when enabled,
+   so the kernel stays config-free and replay/recovery reconstruct the fact from the
+   signal log.
+3. **Engine** (`HandleEspTerminalFailureV1`): a terminal ESP failure is defanged into an
+   `esp_failure_advisory` (`advisoryReason=esp_failure_defanged_continueanyway_observation`)
+   when ALL hold: profile allows Continue-Anyway, observation opt-in observed,
+   `AccountSetupEnteredUtc == null`, and the scenario is Classic/Unknown with
+   `PreProvisioningSide=None` and no registry self-deploying marker (WhiteGlove /
+   SelfDeploying / DevicePreparation have no interactive user to press the button).
+4. **Observation window**: the shared `AdvisoryCompletion` deadline is armed with a
+   **60-minute** window (classic advisory: 30) — long enough for failure screen →
+   Continue anyway → sign-in → desktop, or a full device-phase re-run via "Try again".
+
+## Detecting the real end
+
+The active-advisory-without-AccountSetup shape is unique to this arming site, so
+`IsContinueAnywayObservationActive(state)` (advisory recorded ∧ not resolved ∧
+AccountSetup never entered) needs no separate opt-in re-check. Three outcomes:
+
+* **Continue anyway** — the DAD-validated real-user desktop arrives (defaultuser0/SYSTEM
+  excluded at the detector) and the Hello gate is satisfied (resolved, or policy
+  explicitly disabled without an observed wizard): `HandleDesktopArrivedV1` completes
+  eagerly through Finalizing (trigger `DesktopArrived:ContinueAnywayObservation`) — the
+  session end reflects the real end, not the window edge. The IME user-session gate is
+  deliberately waived: it anchors on `AccountSetupEnteredUtc`, which is structurally
+  absent here. The terminal `enrollment_complete` carries `espSoftFailure=true`,
+  `completionSource=continue_anyway_observation` and the failed category; the backend
+  stamps `EspSoftFailure`/`CompletionSource` on the Sessions row (+ SessionsIndex
+  mirror) and the web renders the amber "with issues" badge plus a detail banner.
+  Sessions completing under the CLASSIC advisory (AccountSetup entered) carry the same
+  marker with `completionSource=continue_anyway_post_accountsetup`.
+* **Try again** — the ESP re-runs the failed category; `DeviceSetupProvisioningComplete`
+  resolves the advisory via the 4910a5a5 recovery hook (`esp_failure_advisory_resolved`)
+  and the session continues on the normal rails, completing clean (no marker).
+* **Nothing** — the window expires: un-defang to `enrollment_failed` with the original
+  `esp_terminal_failure` reason and
+  `advisoryReason=continue_anyway_observation_window_expired_without_completion_evidence`;
+  `LastFailureTrigger=EspTerminalFailure` keeps the likely-stuck app promotion on. Every
+  observed session therefore still gets a terminal verdict ≤ ~60 min after the failure.
+
+Deadline-fire edge cases mirror the classic paths: desktop + unknown Hello promotes to
+`AwaitingHello` with the HelloSafety window (never-observed promote), and enforcement
+progress since arming re-arms the window (30-min re-arm, convergent) instead of failing a
+device mid-"Try again".
+
+## Status semantics
+
+No new terminal status: the session is `Succeeded` + `EspSoftFailure=true` (user
+decision 2026-08-08). Success-rate conventions stay untouched; honesty lives in the
+badge, the detail banner, and ANALYZE-ESP-004 (which continues to fire on the un-defang
+expiry path via `mayHaveContinuedAnyway`).
+
+# Examples
+
+Session `53d1e9f6-4eef-4f3b-a999-c63e96c5148d` (tenant c9787ba2, 2026-08-07) is the
+motivating shape: DeviceSetup/Certificates failed at the 30-min wall while "Encompass
+Hybrid Installer" was still installing, AccountSetup 0/5, Continue-Anyway allowed —
+hard-failed under the default semantics.
+
+Tests: `ContinueAnywayObservationTests` (DecisionCore.Tests) — defang, 60-min arming,
+opt-out/scenario regressions, eager desktop completion with marker, clean-completion
+marker absence, expiry un-defang, hello promotes, category recovery.
+
+# Citations
+
+* `src/Shared/AutopilotMonitor.DecisionCore/Engine/DecisionEngine.Edge.cs` — arming site,
+  observation branches in `HandleAdvisoryCompletionDeadlineFired`,
+  `IsContinueAnywayObservationActive`.
+* `src/Shared/AutopilotMonitor.DecisionCore/Engine/DecisionEngine.Classic.cs` —
+  `HandleDesktopArrivedV1` observation completion, `BuildEnrollmentCompleteEffect`
+  soft-failure marker.
+* `src/Backend/AutopilotMonitor.Functions/Services/EventIngestProcessor.Classification.cs`
+  / `TableStorageService.Sessions.cs` — session stamping + index mirror.
+* `src/Web/autopilot-monitor-web/components/SessionStatusBadge.tsx` — amber badge.

docs/index.md

@@ -23,6 +23,7 @@ https://docs.autopilotmonitor.com (separate repository).
 * [Gather Rule Debug Log](agent/gather-rule-debug-log.md) - EnableGatherRuleDebugLog (ConfigVersion 36): the per-tenant local trace file that explains every gather-rule evaluation decision — scope skips, on_change suppression, empty collector results, logparser position/parse-mode outcomes — for rules that produce no timeline events.
 * [MDM Reboot Coalescing](agent/mdm-reboot-coalescing.md) - Attributing the mid-ESP coalesced reboot ("second sign-in") to the device-assigned policy URIs that forced it: aggregated DM-Enterprise 2800 observations (neutral, flush-time watermark) + the ANALYZE-ESP-005 advisory gated on an actually-observed reboot.
 * [Secure Boot CA-2023 Detection](agent/secureboot-ca2023-detection.md) - Why ANALYZE-SEC-001 v3 reads the UEFI db/KEK variables directly instead of trusting the SecureBoot\Servicing registry, and how the one-sided uefiCa2023FirmwareConfirmed marker plus a not_exists precondition give absence-tolerant suppression across old and new agents.
+* [Continue-Anyway Observation Mode](agent/continue-anyway-observation.md) - EnableEspContinueAnywayObservation (ConfigVersion 37, operator-set): Device-phase ESP terminal failures on Continue-Anyway profiles get a bounded 60-min observation instead of an immediate hard fail; the DAD-validated real-user desktop + Hello gate detect the true end, the session completes Succeeded + EspSoftFailure (amber "with issues" badge), expiry un-defangs to the original esp_terminal_failure.
 * [ESP Policy-Provider Stall Detection](agent/esp-policy-provider-stall.md) - The EnrollmentStatusTracking CSP's no-timeout Setup/Apps wait is keyed to the Intune-registered "Sidecar" provider by NAME: co-management's "ConfigMgr" parks the user ESP at "Apps (Identifying)" for days even with TrackingPoliciesCreated=1 (issue #106); the always-on two-arm esp_policy_provider_stalled tripwire (provider_incomplete | sidecar_provider_missing) and the SOFTWARE\Microsoft\Windows\Autopilot gather-rule allowlist entry that surface it.
 
 # Rules

docs/log.md

@@ -1,5 +1,11 @@
 # Log
 
+## 2026-08-10
+
+* **Update**: `rules/gather-rule-guardrails.md` — WMI matcher generalized from literal prefix matching to shape parsing: `SELECT <* | property-list> FROM <class>` is allowed when the class is derived from the existing `SELECT * FROM <class>` entries (a projection is a strict subset; motivated by `SELECT BatteryStatus FROM Win32_Battery` + `on_change` battery polling that `SELECT *` defeats via fluctuating charge %). Same algorithm mirrored agent/portal/MCP; guardrails.json unchanged. Portal now also gates enabling on the client-side validation (invalid-target custom rules create disabled, cannot be toggled on, force-disable on invalid edit, "Blocked on devices" badge) — UX only, the agent guard stays the boundary. Gather-rule Author is now stamped from the creator's JWT (display name → UPN) at create and immutable on update.
+
+* **New**: `agent/continue-anyway-observation.md` — Continue-Anyway observation mode (`EnableEspContinueAnywayObservation`, ConfigVersion 37, operator-set): Device-phase ESP terminal failures on Continue-Anyway profiles are defanged into a 60-min observation window (Classic/Unknown, non-pre-prov only); real-user desktop + Hello gate complete the session as Succeeded + `EspSoftFailure`/`CompletionSource` (amber "with issues" badge + detail banner), category recovery resolves clean, expiry un-defangs to the original `esp_terminal_failure`. Registered in the Agent index section. Related backend fix (no doc): the auto-analyze enqueue now carries a 30 s `visibilityTimeout` so rule evaluation no longer races the agent's terminal flush (ANALYZE-ESP-004's `{{appName}}` placeholder).
+
 ## 2026-08-07 (later)
 
 * **Update**: `backend/cloudpc-device-validation.md` — third `IsCloudPc` transport rail: the pre-auth distress channel. `DistressReport.IsCloudPc` (CloudPcDetector verdict captured at `DistressReporter` construction) → `DistressReportEntry`/table column → "Devices Not Registered" aggregation (sticky-true per serial, same OR semantics as the Sessions merge) → portal Cloud PC badge + actionable hint ("enable Windows 365 Cloud PC Validation + W365CloudPcValidation add-on"). Motivated by the first W365 field run: the rejected Cloud PC showed up as anonymous "Microsoft Corporation / Virtual Machine" with no clue that the fix is the Cloud PC validator, not Autopilot registration. UNVERIFIED like all distress fields; ≤18 bytes, fits the 1536-byte cap.

docs/rules/gather-rule-guardrails.md

@@ -7,14 +7,14 @@ description: >-
   guard rather than trusting rule.Target.
 resource: src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Telemetry/Gather
 tags: [gather-rules, security, guardrails, agent]
-timestamp: 2026-07-22
+timestamp: 2026-08-10
 ---
 
 # Concept
 
 A gather rule is authored by a **tenant administrator** through `POST /api/rules/gather` and executed by the agent as **SYSTEM** on an enrolling device. The backend does not validate `collectorType`, `target`, or `parameters` at all — it only enforces scope/emit-mode fields, a 1 MB body cap, and that the rule lands in the caller's own tenant partition. Everything that decides *what a rule may reach* is therefore enforced **on the agent**, in `GatherRuleGuards`.
 
-This is the security boundary. The portal's `ValidationIndicator` is a convenience: it does not block submit, and the API is reachable directly.
+This is the security boundary. The portal's `ValidationIndicator` is a convenience — the API stays reachable directly and the backend stays validation-free — but since 2026-08 the portal additionally **gates enabling** on the same client-side validation: a custom rule whose target fails validation cannot be toggled on, is created `enabled=false`, and an edit that makes the target invalid force-disables the rule (`targetBlocked` in `app/gather-rules/types.ts`). An enabled rule with an invalid target shows a "Blocked on devices" badge. All of this is UX, not security — the agent guard remains the enforcement point.
 
 # Schema
 
@@ -27,7 +27,7 @@ item in `AutopilotMonitor.Agent.V2.Core.csproj`) and regenerated into the portal
 | --- | --- | --- |
 | `IsRegistryPathAllowed` | `registryPrefixes` | prefix, boundary `\` or end |
 | `IsFilePathAllowed` | `filePrefixes` | `Path.GetFullPath` first, then prefix + boundary |
-| `IsWmiQueryAllowed` | `wmiQueryPrefixes` | prefix, boundary whitespace or end |
+| `IsWmiQueryAllowed` | `wmiQueryPrefixes` | `SELECT <* \| property-list> FROM <class>`; class = case-insensitive equality against classes derived from the `SELECT * FROM <class>` entries, boundary whitespace or end (trailing `WHERE` allowed). Non-canonical entries (none today) fall back to literal prefix + boundary |
 | `IsCommandAllowed` | `allowedCommands` | **exact** match, trimmed — never prefix |
 | `IsEventLogChannelAllowed` | `eventLogChannels` | prefix, boundary `/` or end |
 
@@ -71,6 +71,15 @@ Two collectors were added to this contract on 2026-07-22 after shipping without
 remote config. The operator using it is already a local admin, so it relaxes the
 allowlist exactly the way unrestricted mode does — hard blocks still apply.
 
+**WMI property projections (2026-08):** `SELECT BatteryStatus FROM Win32_Battery` is
+allowed because `SELECT * FROM Win32_Battery` is — a projection exposes a strict subset
+of the allowed class. The matcher parses the query shape instead of literal-prefix
+matching; the allowed-class set is derived at load time from the existing
+`wmiQueryPrefixes` entries, so `rules/guardrails.json` needs no new key. The same
+algorithm is mirrored in the portal (`utils/guardValidation.ts`) and the MCP server
+(`rule-validation.ts`); agents older than the change still block projections with a
+`security_warning` until they self-update.
+
 # Citations
 
 * `src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Telemetry/Gather/GatherRuleGuards.cs`

scripts/Bootstrap/Install-AutopilotMonitor-dev.ps1

@@ -1,523 +0,0 @@
-<#
-.SYNOPSIS
-    DEV bootstrap script (v2.3-dev) to deploy and start the Autopilot Monitor agent.
-
-.DESCRIPTION
-    DEV VARIANT of Install-AutopilotMonitor.ps1 for validating the v2.3 Cloud-PC-relax
-    on Windows 365 before promoting it to the production script. Differences to v2.2:
-      - Cloud-PC-relax (see below) as a second trigger for the guard 2+3 exemption
-      - Raw OutOfBoxExperienceState value is logged (diagnosis)
-      - Logs to bootstrap_agent_dev.log so runs never interleave with the prod script
-      - Testable structure: guard/relax logic lives in functions; the script only
-        executes when run directly. Dot-sourcing it (as the Pester suite in
-        Install-AutopilotMonitor-dev.Tests.ps1 does) loads the functions without
-        side effects.
-
-    Designed to be deployed via Intune as a PowerShell Script during Autopilot.
-    Runs very early in the enrollment process (first Intune action) and:
-      1. Runs pre-flight guards to skip productive devices and ghost re-installs.
-         A relax exemption allows install when a single freshly-created profile
-         appears DURING OOBE (Windows Backup for Organizations restore) or on a
-         freshly provisioned Windows 365 Cloud PC.
-      2. Downloads the monitoring agent ZIP from Azure Blob Storage
-      3. Verifies integrity via SHA-256 hash from the version manifest (mandatory)
-      4. Extracts the agent into %ProgramData%\AutopilotMonitor\Agent
-      5. Runs the agent in --install mode (registers Scheduled Task, spawns the runtime)
-      6. Verifies the runtime process actually launched
-    Agent self-destructs when enrollment completes.
-
-    Relax flow (guards 2+3): a real user profile (or a real LastLoggedOnUser)
-    normally means "productive -> skip". Two legitimate flows break that assumption:
-
-    (1) Windows Backup for Organizations: one user signs in + restores DURING OOBE.
-        Trigger: the OS positively reports OOBE InProgress (Test via WinRT
-        SystemSetupInfo.OutOfBoxExperienceState).
-
-    (2) Windows 365 Cloud PC: W365 provisions headless -- OOBE and Intune enrollment
-        complete before any user ever connects, so OutOfBoxExperienceState is NEVER
-        InProgress when this script runs. The assigned user's profile is created at
-        first connect, typically seconds before IME pulls this script. Trigger: the
-        device positively identifies as a Cloud PC via TWO local markers (both
-        required; captured from a real Cloud PC -- local WMI reports only the
-        generic 'Virtual Machine' model): HKLM\SOFTWARE\Microsoft\Windows365
-        registry key AND the CloudManagedDesktopExtension service.
-
-    Either trigger relaxes guards 2+3 ONLY when additionally:
-      (a) there is exactly ONE real user profile, AND
-      (b) that profile was created within $OobeProfileMaxAgeMinutes minutes.
-    The Cloud PC trigger additionally exempts guard 4 (uptime window) -- see
-    .PARAMETER MaxBootstrapWindowHours for why boot uptime is meaningless there.
-    Any miss keeps the original skip (SKIP-safe: we never install on uncertainty).
-    On productive devices the profile is old, so (b) keeps skipping correctly.
-    Known accepted edge: a W365 Frontline shared-mode device can legitimately show a
-    single fresh profile while productive; a mis-install there is one-time (agent
-    finds enrollment complete and self-destructs; Deployed marker prevents repeats).
-
-.PARAMETER AgentDownloadUrl
-    URL to download the agent ZIP from. Defaults to the production download endpoint;
-    override only for parallel lab/dev assignments that point at a separate pre-release ZIP.
-
-.PARAMETER VersionJsonName
-    Filename of the integrity manifest in the same blob container as the agent ZIP.
-    Defaults to "version.json"; override only for parallel lab/dev manifests.
-
-.PARAMETER MaxBootstrapWindowHours
-    Maximum device uptime (hours) within which the bootstrap is still considered
-    valid. Devices booted more than this many hours ago are skipped because we no
-    longer trust their OOBE state. Default: 12.
-    Cloud PC note: NOT applied when the Cloud-PC relax is active. A Cloud PC runs
-    headless for days between provisioning and the user's first connect, and the
-    phase worth monitoring (Account Setup) only starts at that first connect --
-    there the fresh-profile window (OobeProfileMaxAgeMinutes) is the time anchor,
-    not boot uptime.
-
-.PARAMETER OobeProfileMaxAgeMinutes
-    Maximum age (minutes) of the single profile that may appear for the relax
-    exemption to apply. The restore/first-connect -> bootstrap gap is always
-    minutes; an older profile is treated as a productive device. Default: 15.
-
-.NOTES
-    - Agent is temporary and auto-removes after enrollment.
-    - All files live under C:\ProgramData\AutopilotMonitor (easy cleanup).
-    - One registry key remains after removal: HKLM\SOFTWARE\AutopilotMonitor\Deployed
-      (prevents ghost re-installs on re-Autopilot of the same device).
-    - Scheduled Task survives reboots during enrollment.
-    - OOBE detection uses Windows.System.Profile.SystemSetupInfo.OutOfBoxExperienceState.
-      Requires Windows 10, version 1809 (introduced in 10.0.17763.0). Older builds lack
-      the class -> state reads as 'Unavailable' -> original (skip) behaviour. Docs:
-      https://learn.microsoft.com/en-us/uwp/api/windows.system.profile.systemsetupinfo
-    - This script MUST remain pure ASCII (no Unicode/UTF-8 special chars).
-      PowerShell 5.1 (IME) reads scripts without BOM as ANSI, corrupting multi-byte chars.
-
-.CHANGELOG
-    2026-08-06  v2.3-dev.3  Guard 4 (uptime window) exempted while the Cloud-PC relax is
-                      active: Cloud PCs run headless for days before the user's first
-                      connect, so boot uptime says nothing about enrollment freshness --
-                      the single-fresh-profile window (< OobeProfileMaxAgeMinutes) already
-                      anchors the relax in time. OOBE-restore keeps the uptime guard.
-    2026-08-06  v2.3-dev.2  Cloud PC detection rebuilt on captured evidence: first field
-                      test proved local WMI reports only Manufacturer='Microsoft
-                      Corporation' + Model='Virtual Machine' (the 'Cloud PC ...' model
-                      string exists only in Intune/Graph). Detection now requires the
-                      HKLM\SOFTWARE\Microsoft\Windows365 registry key AND the
-                      CloudManagedDesktopExtension service (both verified present on a
-                      real Cloud PC); raw identity + marker states are always logged.
-    2026-08-05  v2.3-dev  DEV variant (-dev.ps1, own log file bootstrap_agent_dev.log)
-                      to validate on Windows 365 before promotion to v2.3:
-                      Cloud-PC-relax -- a positively identified Cloud PC
-                      (Win32_ComputerSystem Manufacturer 'Microsoft Corporation' +
-                      Model 'Cloud PC*') now triggers the guard 2+3 relax alongside
-                      OOBE InProgress; the single-fresh-profile conditions are
-                      unchanged. Raw OutOfBoxExperienceState is now logged.
-                      Refactored into functions with a dot-source entry guard so the
-                      Pester suite (*.Tests.ps1) can load the logic without executing
-                      the bootstrap; behaviour when run directly is unchanged.
-    2026-07-20  v2.2  Switched default download endpoint from the legacy blob
-                      (autopilotmonitor.blob.core.windows.net) to the Front Door alias
-                      https://download.autopilotmonitor.com/agent/ (route /agent/*).
-                      version.json is still derived from the same container.
-    2026-06-25  v2.1  OOBE-relax: detect genuine OOBE via WinRT
-                      SystemSetupInfo.OutOfBoxExperienceState and exempt guards 2+3 when a
-                      single freshly-created profile appears in OOBE (Windows Backup for
-                      Organizations). SKIP-safe when the API is absent (< 1809).
-                      Removed legacy Content-MD5 fallback; SHA-256 manifest verification is
-                      now mandatory (no hash -> bootstrap aborts, no unverified install).
-    2026-05-09  v2.0  Generic bootstrap: agent owns its own defaults (e.g. 600 s
-                      TenantId-wait), so the script only calls `--install` plain.
-                      Hardened post-install with a 10 s runtime-process verify.
-    (older entries: see Install-AutopilotMonitor.ps1)
-#>
-
-[CmdletBinding()]
-param(
-    [Parameter(Mandatory = $false)]
-    [string]$AgentDownloadUrl = "https://download.autopilotmonitor.com/agent/AutopilotMonitor-Agent.zip",
-
-    [Parameter(Mandatory = $false)]
-    [string]$VersionJsonName = "version.json",
-
-    [Parameter(Mandatory = $false)]
-    [int]$MaxBootstrapWindowHours = 12,
-
-    [Parameter(Mandatory = $false)]
-    [int]$OobeProfileMaxAgeMinutes = 15
-)
-
-# Script version (bump on meaningful changes; see .CHANGELOG above)
-$ScriptVersion = "2.3-dev.3"
-
-# Configuration - Everything in ProgramData for easy cleanup
-$AgentBasePath = "$env:ProgramData\AutopilotMonitor"
-$AgentBinPath = "$AgentBasePath\Agent"
-$AgentLogPath = "$AgentBasePath\Logs"
-$LogFile = "$AgentLogPath\bootstrap_agent_dev.log"
-
-# Write-Host (NOT Write-Output): log calls happen inside functions that return
-# decision objects; Write-Output would leak every log line into those return
-# values and turn them into arrays. Write-Host goes to the console host only.
-function Write-Log {
-    param([string]$Message)
-    $timestamp = Get-Date -Format "yyyy-MM-dd HH:mm:ss"
-    $logMessage = "[$timestamp] $Message"
-    Write-Host $logMessage
-    Add-Content -Path $LogFile -Value $logMessage -ErrorAction SilentlyContinue
-}
-
-# Raw OOBE state via the documented WinRT API
-# Windows.System.Profile.SystemSetupInfo.OutOfBoxExperienceState.
-# Returns the state string (InProgress/Completed/NotStarted). An older OS without the
-# class (< Win10 1809 / 10.0.17763.0) or any error returns 'Unavailable'
-# -> SKIP-safe (we never relax a guard on uncertainty).
-# Docs: https://learn.microsoft.com/en-us/uwp/api/windows.system.profile.systemsetupinfo
-function Get-OobeState {
-    try {
-        $null = [Windows.System.Profile.SystemSetupInfo, Windows.System.Profile, ContentType = WindowsRuntime]
-        return ([Windows.System.Profile.SystemSetupInfo]::OutOfBoxExperienceState).ToString()
-    }
-    catch {
-        return 'Unavailable'
-    }
-}
-
-# Positive Windows 365 Cloud PC detection.
-# Evidence captured 2026-08-06 on a real, freshly provisioned W365 Enterprise Cloud PC:
-#   - Win32_ComputerSystem is GENERIC there: Manufacturer='Microsoft Corporation',
-#     Model='Virtual Machine'. The 'Cloud PC Enterprise ...' model string exists only
-#     in the Intune/Graph inventory, NOT in local WMI/SMBIOS -- it cannot be used here.
-#   - HKLM:\SOFTWARE\Microsoft\Windows365 exists.
-#   - Service 'CloudManagedDesktopExtension' (Microsoft Cloud Managed Desktop
-#     Extension, the W365 management agent that runs ON Cloud PCs) is installed.
-# BOTH markers are required (AND): each alone has plausible look-alikes (W365-Boot
-# physical clients may carry Windows365 policy state; the MCMD agent family also
-# served other managed-desktop offerings). Raw identity and both marker states are
-# always logged for field diagnosis; any error returns $false (SKIP-safe).
-function Test-IsCloudPc {
-    try {
-        $cs = Get-CimInstance Win32_ComputerSystem -ErrorAction Stop
-        Write-Log "Computer identity: Manufacturer='$($cs.Manufacturer)' Model='$($cs.Model)'."
-    }
-    catch { Write-Log "INFO: computer identity query failed: $($_.Exception.Message)" }
-
-    try {
-        $regMarker = Test-Path 'HKLM:\SOFTWARE\Microsoft\Windows365'
-        $svcMarker = $null -ne (Get-Service -Name 'CloudManagedDesktopExtension' -ErrorAction SilentlyContinue)
-        Write-Log "Cloud PC markers: Windows365 registry key=$regMarker; CloudManagedDesktopExtension service=$svcMarker."
-        return ($regMarker -and $svcMarker)
-    }
-    catch {
-        Write-Log "INFO: Cloud PC marker check failed: $($_.Exception.Message)"
-        return $false
-    }
-}
-
-# Collects real (non-special) user profiles as full paths (WMI Special flag + filesystem).
-# NOTE: callers MUST wrap the result in @(...). PowerShell unrolls function output, so an
-# empty result returns $null and a single path returns a scalar string, whose [0] indexer
-# would yield its first char ('C') instead of the full path.
-function Get-RealUserProfilePaths {
-    $excludePattern = '^(defaultuser\d*|Public|Default( User)?|All Users|WDAGUtilityAccount)$'
-    @(
-        try {
-            Get-CimInstance Win32_UserProfile -ErrorAction Stop |
-                Where-Object { -not $_.Special -and $_.LocalPath -like 'C:\Users\*' } |
-                ForEach-Object { $_.LocalPath }
-        } catch { Write-Log "INFO: WMI profile query failed, continuing with filesystem check." }
-
-        (Get-ChildItem 'C:\Users' -Directory -ErrorAction SilentlyContinue).FullName
-    ) | Where-Object { $_ -and (Split-Path $_ -Leaf) -notmatch $excludePattern } |
-        Select-Object -Unique
-}
-
-# Relax: the ONLY exemption to guards 2+3. Two triggers (see .DESCRIPTION):
-#   (1) OOBE InProgress          -> Windows Backup for Organizations restore
-#   (2) Cloud PC identity        -> Windows 365 (headless OOBE, profile at first connect)
-# Either trigger additionally requires: exactly one real profile AND that profile
-# created < $OobeProfileMaxAgeMinutes minutes ago. Any miss keeps the original skip.
-# Returns an object with: Active (bool), OobeState (string), IsCloudPc (bool).
-function Get-RelaxDecision {
-    param(
-        [string[]]$ProfilePaths,
-        [int]$OobeProfileMaxAgeMinutes
-    )
-    if ($null -eq $ProfilePaths) { $ProfilePaths = @() }
-
-    $relaxActive = $false
-    $oobeState = Get-OobeState
-    $oobeInProgress = ($oobeState -eq 'InProgress')
-    $isCloudPc = Test-IsCloudPc
-    if ($isCloudPc) {
-        Write-Log "Cloud PC detected (OOBE state: $oobeState)."
-    }
-
-    $relaxTrigger = $null
-    if ($oobeInProgress) {
-        $relaxTrigger = "OOBE InProgress (Windows Backup restore case)"
-    } elseif ($isCloudPc) {
-        $relaxTrigger = "Cloud PC (headless provisioning, OOBE state: $oobeState)"
-    }
-
-    if ($ProfilePaths.Count -eq 1 -and $relaxTrigger) {
-        try {
-            $created = (Get-Item -LiteralPath $ProfilePaths[0] -Force -ErrorAction Stop).CreationTimeUtc
-            $ageMin = ((Get-Date).ToUniversalTime() - $created).TotalMinutes
-            if ($ageMin -ge 0 -and $ageMin -lt $OobeProfileMaxAgeMinutes) {
-                $relaxActive = $true
-                Write-Log ("Relax active: {0} + single profile created {1:N1}min ago (< {2}min). Guards 2+3 relaxed." -f $relaxTrigger, $ageMin, $OobeProfileMaxAgeMinutes)
-            } else {
-                Write-Log ("Relax NOT applied: {0} + single profile '{1}', but profile created {2:N1}min ago (created {3:u}) is outside the [0, {4})min window." -f $relaxTrigger, (Split-Path $ProfilePaths[0] -Leaf), $ageMin, $created, $OobeProfileMaxAgeMinutes)
-            }
-        } catch { Write-Log "INFO: profile age check failed; relax not applied. $($_.Exception.Message)" }
-    } elseif ($ProfilePaths.Count -eq 1) {
-        Write-Log "Relax NOT applied: single profile present but neither OOBE InProgress nor Cloud PC (OOBE state: $oobeState)."
-    } elseif ($ProfilePaths.Count -gt 1) {
-        Write-Log ("Relax NOT applied: {0} real profiles present (relax requires exactly one). OOBE state: {1}; CloudPC: {2}." -f $ProfilePaths.Count, $oobeState, $isCloudPc)
-    }
-
-    return [pscustomobject]@{
-        Active    = $relaxActive
-        OobeState = $oobeState
-        IsCloudPc = $isCloudPc
-    }
-}
-
-# Runs pre-flight guards 1-5 and returns the install decision WITHOUT side effects
-# beyond logging. Returns an object with:
-#   Install (bool), ReasonCode (null | AlreadyDeployed | ProfilesFound |
-#   LastLoggedOnUser | UptimeExceeded | AgentPresent), RelaxActive (bool).
-function Get-BootstrapDecision {
-    param(
-        [int]$MaxBootstrapWindowHours,
-        [int]$OobeProfileMaxAgeMinutes,
-        [string]$AgentBinPath
-    )
-
-    # Guard 1: Agent was already deployed on this device (registry marker survives self-destruct)
-    $deployed = (Get-ItemProperty -Path 'HKLM:\SOFTWARE\AutopilotMonitor' -Name 'Deployed' -ErrorAction SilentlyContinue).Deployed
-    if ($deployed) {
-        Write-Log "SKIP: Agent was previously deployed at $deployed."
-        return [pscustomobject]@{ Install = $false; ReasonCode = 'AlreadyDeployed'; RelaxActive = $false }
-    }
-
-    # The @() wrap is required, see NOTE on Get-RealUserProfilePaths.
-    $profilePaths = @(Get-RealUserProfilePaths)
-    $relax = Get-RelaxDecision -ProfilePaths $profilePaths -OobeProfileMaxAgeMinutes $OobeProfileMaxAgeMinutes
-
-    # Guard 2: No real user profile should exist yet (primary productive-device guard),
-    # unless the relax signature holds.
-    if ($profilePaths.Count -gt 0 -and -not $relax.Active) {
-        $details = (($profilePaths | ForEach-Object {
-            $leaf = Split-Path $_ -Leaf
-            try {
-                $c = (Get-Item -LiteralPath $_ -Force -ErrorAction Stop).CreationTimeUtc
-                "{0} (created {1:u}, {2:N1}min ago)" -f $leaf, $c, ((Get-Date).ToUniversalTime() - $c).TotalMinutes
-            } catch { "$leaf (age unknown)" }
-        }) | Select-Object -First 3) -join '; '
-        Write-Log "SKIP: Real user profile(s) found ($details). OOBE state: $($relax.OobeState). CloudPC: $($relax.IsCloudPc). Device appears productive."
-        return [pscustomobject]@{ Install = $false; ReasonCode = 'ProfilesFound'; RelaxActive = $false }
-    }
-
-    # Guard 3: LastLoggedOnUser - during Device ESP no real user has logged on yet
-    # (same relax exemption: the restoring/first-connect user signs in legitimately).
-    $lastLoggedOnUser = (Get-ItemProperty -Path 'HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Authentication\LogonUI' -Name 'LastLoggedOnUser' -EA SilentlyContinue).LastLoggedOnUser
-    if ($lastLoggedOnUser -and $lastLoggedOnUser -notmatch 'defaultuser\d*' -and -not $relax.Active) {
-        Write-Log "SKIP: LastLoggedOnUser found ($lastLoggedOnUser). Device appears productive."
-        return [pscustomobject]@{ Install = $false; ReasonCode = 'LastLoggedOnUser'; RelaxActive = $false }
-    }
-
-    # Guard 4: Bootstrap window check - device uptime must be within accepted OOBE window.
-    # Sleep/standby does not reset uptime, only real boot/restart does.
-    # Cloud PC exemption (v2.3-dev.3): a Cloud PC runs headless from provisioning until the
-    # user's first connect -- often days -- and the phase worth monitoring (Account Setup)
-    # only STARTS at that first connect. Boot uptime is therefore meaningless there; the
-    # single-fresh-profile window of the relax (< $OobeProfileMaxAgeMinutes min) is the time
-    # anchor instead. The OOBE-restore trigger keeps the original uptime guard (physical
-    # device, uptime is meaningful).
-    $lastBoot = (Get-CimInstance Win32_OperatingSystem).LastBootUpTime
-    $uptimeHours = ((Get-Date) - $lastBoot).TotalHours
-    Write-Log "Device uptime: $([int]$uptimeHours)h (last boot: $lastBoot)"
-    if ($uptimeHours -gt $MaxBootstrapWindowHours) {
-        if ($relax.Active -and $relax.IsCloudPc) {
-            Write-Log "Uptime $([int]$uptimeHours)h exceeds the ${MaxBootstrapWindowHours}h window, but the Cloud PC relax is active (fresh first-connect profile is the time anchor). Continuing."
-        } else {
-            Write-Log "SKIP: Device uptime is $([int]$uptimeHours)h. OOBE state is older than accepted bootstrap window of ${MaxBootstrapWindowHours}h."
-            return [pscustomobject]@{ Install = $false; ReasonCode = 'UptimeExceeded'; RelaxActive = $relax.Active }
-        }
-    }
-
-    # Guard 5: Is the agent already installed? (leftover from previous run)
-    if (Test-Path $AgentBinPath) {
-        $existingAgent = Get-ChildItem -Path $AgentBinPath -Filter "AutopilotMonitor.Agent.exe" -ErrorAction SilentlyContinue
-        if ($existingAgent) {
-            Write-Log "SKIP: Agent already installed at $($existingAgent.FullName)."
-            return [pscustomobject]@{ Install = $false; ReasonCode = 'AgentPresent'; RelaxActive = $relax.Active }
-        }
-    }
-
-    return [pscustomobject]@{ Install = $true; ReasonCode = $null; RelaxActive = $relax.Active }
-}
-
-# Main bootstrap flow. Only invoked when the script is run directly (see entry guard
-# at the bottom); never on dot-source.
-function Invoke-Bootstrap {
-    try {
-        New-Item -Path $AgentBasePath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
-        New-Item -Path $AgentBinPath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
-        New-Item -Path $AgentLogPath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
-
-        Write-Log "===== Autopilot Monitor Bootstrap Started ====="
-        Write-Log "Bootstrap script version: v$ScriptVersion"
-
-        $decision = Get-BootstrapDecision -MaxBootstrapWindowHours $MaxBootstrapWindowHours -OobeProfileMaxAgeMinutes $OobeProfileMaxAgeMinutes -AgentBinPath $AgentBinPath
-        if (-not $decision.Install) {
-            exit 0
-        }
-
-        Write-Log "Pre-flight checks passed -- no prior deployment, no productive-device signals, within bootstrap window (relaxActive=$($decision.RelaxActive))"
-
-        # Download and extract agent binaries
-        $agentExePath = Join-Path $AgentBinPath "AutopilotMonitor.Agent.exe"
-
-        if (Test-Path $agentExePath) {
-            Write-Log "Agent already installed at $agentExePath"
-        }
-        else {
-            Write-Log "Downloading agent from $AgentDownloadUrl..."
-
-            try {
-                # Derive manifest URL from the agent download URL (same blob container)
-                $versionJsonUrl = $AgentDownloadUrl -replace '[^/]+$', $VersionJsonName
-                $expectedSha256 = $null
-                $manifestAttempts = 3
-                $manifestDelays = @(1, 2)   # short backoff between attempts; do not block enrollment
-
-                for ($attempt = 1; $attempt -le $manifestAttempts; $attempt++) {
-                    try {
-                        Write-Log "Fetching $VersionJsonName (attempt ${attempt}/${manifestAttempts}) from $versionJsonUrl for integrity verification..."
-                        $versionJsonResponse = Invoke-RestMethod -Uri $versionJsonUrl -UseBasicParsing -TimeoutSec 5 -ErrorAction Stop
-                        if ($versionJsonResponse.sha256) {
-                            $expectedSha256 = $versionJsonResponse.sha256.ToLowerInvariant()
-                            Write-Log "SHA-256 hash from manifest: $expectedSha256 (version: $($versionJsonResponse.version))"
-                        } else {
-                            Write-Log "WARNING: Manifest has no sha256 field - integrity cannot be verified."
-                        }
-                        break   # manifest reached (with or without hash); retrying would not help
-                    }
-                    catch {
-                        Write-Log "WARNING: Could not fetch integrity manifest (attempt $attempt): $($_.Exception.Message)"
-                        if ($attempt -lt $manifestAttempts) {
-                            Start-Sleep -Seconds $manifestDelays[$attempt - 1]
-                        }
-                    }
-                }
-
-                $zipPath = Join-Path $env:TEMP "AutopilotMonitor-Agent.zip"
-                $maxDownloadAttempts = 3
-                $downloadAttempt = 0
-
-                do {
-                    $downloadAttempt++
-                    try {
-                        Write-Log "Download attempt ${downloadAttempt}/${maxDownloadAttempts}"
-                        $downloadStopwatch = [System.Diagnostics.Stopwatch]::StartNew()
-                        Invoke-WebRequest `
-                            -Uri $AgentDownloadUrl `
-                            -OutFile $zipPath `
-                            -UseBasicParsing `
-                            -TimeoutSec 30 `
-                            -ErrorAction Stop
-                        $downloadStopwatch.Stop()
-                        $downloadSeconds = [math]::Round($downloadStopwatch.Elapsed.TotalSeconds, 1)
-                        Write-Log "Downloaded agent to $zipPath (took ${downloadSeconds}s)"
-                        break
-                    }
-                    catch {
-                        if ($downloadAttempt -ge $maxDownloadAttempts) {
-                            throw
-                        }
-
-                        $retryDelaysInSeconds = @(2, 4, 8)
-                        $retryDelaySeconds = $retryDelaysInSeconds[$downloadAttempt - 1]
-                        Write-Log "Download failed (attempt $downloadAttempt): $($_.Exception.Message). Retrying in ${retryDelaySeconds}s..."
-                        Start-Sleep -Seconds $retryDelaySeconds
-                    }
-                } while ($downloadAttempt -lt $maxDownloadAttempts)
-
-                # SHA-256 from the version manifest is the only accepted integrity proof.
-                # No hash -> refuse to install an unverified ZIP (SKIP-safe security posture).
-                if (-not $expectedSha256) {
-                    throw "No SHA-256 hash available from manifest - refusing to install unverified agent ZIP."
-                }
-                $actualSha256 = (Get-FileHash $zipPath -Algorithm SHA256).Hash.ToLowerInvariant()
-                Write-Log "Validating SHA-256 hash against manifest"
-                if ($actualSha256 -ne $expectedSha256) {
-                    throw "SHA-256 integrity check FAILED. Expected='$expectedSha256', Actual='$actualSha256'. Download may be tampered or corrupted."
-                }
-                Write-Log "SHA-256 integrity check passed"
-
-                Expand-Archive -Path $zipPath -DestinationPath $AgentBinPath -Force
-                Write-Log "Extracted agent to $AgentBinPath"
-
-                Remove-Item $zipPath -Force -ErrorAction SilentlyContinue
-                Write-Log "Cleaned up temporary files"
-
-                if (-not (Test-Path $agentExePath)) {
-                    throw "Agent executable not found after extraction at $agentExePath"
-                }
-
-                Write-Log "Agent installation completed successfully"
-            }
-            catch {
-                Write-Log "ERROR downloading/extracting agent: $($_.Exception.Message)"
-                throw
-            }
-        }
-
-        Write-Log "Calling agent install mode (--install)..."
-        & $agentExePath --install
-        $installExitCode = $LASTEXITCODE
-        if ($installExitCode -ne 0) {
-            throw "Agent install failed with exit code $installExitCode"
-        }
-        Write-Log "Agent install mode completed successfully"
-
-        $runtimeProcessName = 'AutopilotMonitor.Agent'
-        $verifyTimeoutSec = 10
-        $verifyDeadline = (Get-Date).AddSeconds($verifyTimeoutSec)
-        $runtimeProc = $null
-        while ((Get-Date) -lt $verifyDeadline) {
-            $runtimeProc = Get-Process -Name $runtimeProcessName -ErrorAction SilentlyContinue | Select-Object -First 1
-            if ($null -ne $runtimeProc) { break }
-            Start-Sleep -Milliseconds 500
-        }
-        if ($null -ne $runtimeProc) {
-            $startedUtc = 'unavailable'
-            try { $startedUtc = $runtimeProc.StartTime.ToUniversalTime().ToString('o') } catch { }
-            Write-Log ("Runtime process verified: name={0}.exe pid={1} startedUtc={2}" -f $runtimeProcessName, $runtimeProc.Id, $startedUtc)
-        } else {
-            Write-Log ("WARNING: Runtime process verification FAILED. Agent --install reported success but no '{0}.exe' process appeared within {1}s. Likely silent block (AV/EDR, AppLocker/WDAC) of the runtime launch. Agent should still come up at next boot via the BootTrigger task. Check Event Viewer > Microsoft > Windows > TaskScheduler/Operational and AV/EDR logs for '{0}.exe'." -f $runtimeProcessName, $verifyTimeoutSec)
-        }
-
-        Write-Log "===== Bootstrap Completed Successfully ====="
-
-        exit 0
-    }
-    catch {
-        Write-Log "===== Bootstrap FAILED ====="
-        Write-Log "ERROR: $($_.Exception.Message)"
-        Write-Log "Stack trace: $($_.ScriptStackTrace)"
-        Write-Log "Please check log file: $LogFile"
-
-        $errMsg = "AutopilotMonitor bootstrap failed: $($_.Exception.Message)"
-        if ($errMsg.Length -gt 2048) { $errMsg = $errMsg.Substring(0, 1045) + '...' }
-        [Console]::Error.WriteLine($errMsg)
-
-        exit 1
-    }
-}
-
-# Entry guard: execute only when run as a script (IME, console). Dot-sourcing
-# (". .\Install-AutopilotMonitor-dev.ps1", as the Pester suite does) only loads
-# the functions above and MUST stay side-effect free.
-if ($MyInvocation.InvocationName -ne '.') {
-    Invoke-Bootstrap
-}

scripts/Bootstrap/Install-AutopilotMonitor.Tests.ps1

@@ -1,5 +1,5 @@
 # Pester tests (Pester 5+) for the bootstrap guard/relax logic in
-# Install-AutopilotMonitor-dev.ps1 (v2.3-dev).
+# Install-AutopilotMonitor.ps1.
 #
 # The script under test is dot-sourced; its entry guard prevents the bootstrap
 # from executing, so loading it here is side-effect free. External state (WMI,
@@ -12,7 +12,7 @@
 # This file MUST remain pure ASCII (PS 5.1 reads BOM-less files as ANSI).
 
 BeforeAll {
-    . (Join-Path $PSScriptRoot 'Install-AutopilotMonitor-dev.ps1')
+    . (Join-Path $PSScriptRoot 'Install-AutopilotMonitor.ps1')
 
     function New-FakeProfile {
         param([string]$Name, [double]$AgeMinutes)
@@ -24,9 +24,9 @@ BeforeAll {
 }
 
 Describe 'Test-IsCloudPc' {
-    # Marker set captured 2026-08-06 from a real W365 Enterprise Cloud PC: local WMI
-    # is generic ('Virtual Machine'), so detection rests on the Windows365 registry
-    # key AND the CloudManagedDesktopExtension service (both required).
+    # Local WMI on a Cloud PC is generic ('Virtual Machine'), so detection rests on
+    # the Windows365 registry key AND the CloudManagedDesktopExtension service
+    # (both required).
     BeforeEach {
         Mock Write-Log { }
         Mock Get-CimInstance { [pscustomobject]@{ Manufacturer = 'Microsoft Corporation'; Model = 'Virtual Machine' } }
@@ -78,21 +78,21 @@ Describe 'Get-RealUserProfilePaths' {
     It 'merges WMI and filesystem views, drops special/system profiles, dedupes' {
         Mock Get-CimInstance {
             @(
-                [pscustomobject]@{ Special = $false; LocalPath = 'C:\Users\LeonGottschalk' }
+                [pscustomobject]@{ Special = $false; LocalPath = 'C:\Users\UserOne' }
                 [pscustomobject]@{ Special = $true; LocalPath = 'C:\Users\systemprofile' }
                 [pscustomobject]@{ Special = $false; LocalPath = 'D:\Profiles\Elsewhere' }
             )
         }
         Mock Get-ChildItem {
             @(
-                [pscustomobject]@{ FullName = 'C:\Users\LeonGottschalk' }
+                [pscustomobject]@{ FullName = 'C:\Users\UserOne' }
                 [pscustomobject]@{ FullName = 'C:\Users\Public' }
                 [pscustomobject]@{ FullName = 'C:\Users\defaultuser0' }
                 [pscustomobject]@{ FullName = 'C:\Users\WDAGUtilityAccount' }
             )
         }
         $result = @(Get-RealUserProfilePaths)
-        $result | Should -Be @('C:\Users\LeonGottschalk')
+        $result | Should -Be @('C:\Users\UserOne')
     }
 
     It 'falls back to the filesystem view when the WMI query fails' {
@@ -132,15 +132,15 @@ Describe 'Get-RelaxDecision' {
             Mock Test-IsCloudPc { $true }
         }
 
-        It 'relaxes for a single fresh profile although OOBE is Completed (W365 field case)' {
-            $p = New-FakeProfile 'LeonGottschalk' 0.6
+        It 'relaxes for a single fresh profile although OOBE is Completed (first connect)' {
+            $p = New-FakeProfile 'CloudPcUser' 0.6
             $d = Get-RelaxDecision -ProfilePaths @($p) -OobeProfileMaxAgeMinutes 15
             $d.Active | Should -BeTrue
             $d.IsCloudPc | Should -BeTrue
         }
 
         It 'does not relax for an old profile (productive Cloud PC)' {
-            $p = New-FakeProfile 'LeonGottschalk' 30
+            $p = New-FakeProfile 'CloudPcUser' 30
             (Get-RelaxDecision -ProfilePaths @($p) -OobeProfileMaxAgeMinutes 15).Active | Should -BeFalse
         }
 
@@ -162,7 +162,7 @@ Describe 'Get-RelaxDecision' {
         }
     }
 
-    Context 'OOBE InProgress (Windows Backup for Organizations, v2.2 behaviour)' {
+    Context 'OOBE InProgress (Windows Backup for Organizations)' {
         It 'relaxes for a single fresh profile during OOBE' {
             Mock Get-OobeState { 'InProgress' }
             $p = New-FakeProfile 'RestoreUser' 3
@@ -218,12 +218,12 @@ Describe 'Get-BootstrapDecision' {
         $d.ReasonCode | Should -Be 'ProfilesFound'
     }
 
-    It 'installs on a freshly provisioned Cloud PC with one fresh profile (W365 field case)' {
-        $script:fakeProfile = New-FakeProfile 'LeonGottschalk' 0.6
+    It 'installs on a freshly provisioned Cloud PC with one fresh profile' {
+        $script:fakeProfile = New-FakeProfile 'CloudPcUser' 0.6
         Mock Get-RealUserProfilePaths { @($script:fakeProfile) }
         Mock Test-IsCloudPc { $true }
         # First-connect user is already visible in LogonUI on W365 -- must not block either.
-        Mock Get-ItemProperty { [pscustomobject]@{ LastLoggedOnUser = 'AzureAD\LeonGottschalk' } } -ParameterFilter { $Path -like '*LogonUI*' }
+        Mock Get-ItemProperty { [pscustomobject]@{ LastLoggedOnUser = 'AzureAD\CloudPcUser' } } -ParameterFilter { $Path -like '*LogonUI*' }
         $d = Get-BootstrapDecision -MaxBootstrapWindowHours 12 -OobeProfileMaxAgeMinutes 15 -AgentBinPath $script:cleanAgentBin
         $d.Install | Should -BeTrue
         $d.RelaxActive | Should -BeTrue
@@ -248,7 +248,7 @@ Describe 'Get-BootstrapDecision' {
         $d.ReasonCode | Should -Be 'UptimeExceeded'
     }
 
-    It 'installs on a Cloud PC despite uptime over the window when the relax is active (guard 4 exemption, v2.3-dev.3)' {
+    It 'installs on a Cloud PC despite uptime over the window when the relax is active (guard 4 exemption)' {
         # W365 reality: provisioned days ago, running headless ever since; the user's
         # first connect just created the profile. Uptime is meaningless here.
         $script:fakeProfile = New-FakeProfile 'FirstConnectUser' 0.6

scripts/Bootstrap/Install-AutopilotMonitor.ps1

@@ -6,8 +6,9 @@
     Designed to be deployed via Intune as a PowerShell Script during Autopilot.
     Runs very early in the enrollment process (first Intune action) and:
       1. Runs pre-flight guards to skip productive devices and ghost re-installs.
-         An OOBE-relax exemption allows install when a single freshly-created profile
-         appears DURING OOBE (the "Windows Backup for Organizations" restore case).
+         A relax exemption allows install when a single freshly-created profile
+         appears DURING OOBE (Windows Backup for Organizations restore) or on a
+         freshly provisioned Windows 365 Cloud PC.
       2. Downloads the monitoring agent ZIP from Azure Blob Storage
       3. Verifies integrity via SHA-256 hash from the version manifest (mandatory)
       4. Extracts the agent into %ProgramData%\AutopilotMonitor\Agent
@@ -15,14 +16,27 @@
       6. Verifies the runtime process actually launched
     Agent self-destructs when enrollment completes.
 
-    OOBE-relax flow (guards 2+3): a real user profile (or a real LastLoggedOnUser)
-    normally means "productive -> skip". Windows Backup for Organizations breaks that
-    by restoring settings and creating ONE profile while still in OOBE. We relax guards
-    2+3 ONLY when all three positives hold, and nothing else:
-      (a) the OS positively reports OOBE InProgress (Test-OobeInProgress), AND
-      (b) there is exactly ONE real user profile, AND
-      (c) that profile was created within $OobeProfileMaxAgeMinutes minutes.
-    Any miss keeps the original skip (SKIP-safe: we never install on uncertainty).
+    Relax flow (guards 2+3): a real user profile (or a real LastLoggedOnUser)
+    normally means "productive -> skip". Two legitimate flows break that assumption:
+
+    (1) Windows Backup for Organizations: one user signs in + restores DURING OOBE.
+        Trigger: the OS positively reports OOBE InProgress.
+
+    (2) Windows 365 Cloud PC: W365 provisions headless -- OOBE and Intune enrollment
+        complete before any user ever connects, so OOBE is never InProgress when
+        this script runs. The assigned user's profile is created at first connect,
+        typically seconds before Intune pulls this script. Trigger: the device
+        positively identifies as a Cloud PC via TWO local markers (both required):
+        the HKLM\SOFTWARE\Microsoft\Windows365 registry key AND the
+        CloudManagedDesktopExtension service.
+
+    Either trigger relaxes guards 2+3 ONLY when additionally:
+      (a) there is exactly ONE real user profile, AND
+      (b) that profile was created within $OobeProfileMaxAgeMinutes minutes.
+    The Cloud PC trigger additionally exempts guard 4 (uptime window) -- see
+    .PARAMETER MaxBootstrapWindowHours. Any miss keeps the original skip
+    (SKIP-safe: we never install on uncertainty). On productive devices the
+    profile is old, so (b) keeps skipping correctly.
 
 .PARAMETER AgentDownloadUrl
     URL to download the agent ZIP from. Defaults to the production download endpoint;
@@ -36,11 +50,14 @@
     Maximum device uptime (hours) within which the bootstrap is still considered
     valid. Devices booted more than this many hours ago are skipped because we no
     longer trust their OOBE state. Default: 12.
+    Not applied while the Cloud PC relax is active: a Cloud PC runs headless for
+    days between provisioning and the user's first connect, so the fresh-profile
+    window (OobeProfileMaxAgeMinutes) is the time anchor there, not boot uptime.
 
 .PARAMETER OobeProfileMaxAgeMinutes
-    Maximum age (minutes) of the single profile that may appear during OOBE for the
-    OOBE-relax exemption to apply. The restore->bootstrap gap is always minutes; an
-    older profile is treated as a productive device. Default: 15.
+    Maximum age (minutes) of the single profile that may appear for the relax
+    exemption to apply. The restore/first-connect -> bootstrap gap is always
+    minutes; an older profile is treated as a productive device. Default: 15.
 
 .NOTES
     - Agent is temporary and auto-removes after enrollment.
@@ -50,12 +67,24 @@
     - Scheduled Task survives reboots during enrollment.
     - OOBE detection uses Windows.System.Profile.SystemSetupInfo.OutOfBoxExperienceState.
       Requires Windows 10, version 1809 (introduced in 10.0.17763.0). Older builds lack
-      the class -> Test-OobeInProgress returns $false -> original (skip) behaviour. Docs:
+      the class -> state reads as 'Unavailable' -> original (skip) behaviour. Docs:
       https://learn.microsoft.com/en-us/uwp/api/windows.system.profile.systemsetupinfo
+    - Guard/relax logic lives in functions; the script only executes when run
+      directly. Dot-sourcing it (as the Pester suite in
+      Install-AutopilotMonitor.Tests.ps1 does) loads the functions without side effects.
     - This script MUST remain pure ASCII (no Unicode/UTF-8 special chars).
       PowerShell 5.1 (IME) reads scripts without BOM as ANSI, corrupting multi-byte chars.
 
 .CHANGELOG
+    2026-08-08  v2.3  Windows 365 Cloud PC support: a positively identified Cloud PC
+                      (Windows365 registry key AND CloudManagedDesktopExtension
+                      service) now triggers the guard 2+3 relax alongside OOBE
+                      InProgress; the single-fresh-profile conditions are unchanged
+                      and guard 4 (uptime window) is exempted for Cloud PCs.
+                      Raw OOBE state and Cloud PC marker states are now logged.
+                      Guard/relax logic refactored into functions with a dot-source
+                      entry guard so the Pester suite can test it without executing
+                      the bootstrap; behaviour when run directly is unchanged.
     2026-07-20  v2.2  Switched default download endpoint from the legacy blob
                       (autopilotmonitor.blob.core.windows.net) to the Front Door alias
                       https://download.autopilotmonitor.com/agent/ (route /agent/*).
@@ -97,7 +126,7 @@ param(
 )
 
 # Script version (bump on meaningful changes; see .CHANGELOG above)
-$ScriptVersion = "2.2"
+$ScriptVersion = "2.3"
 
 # Configuration - Everything in ProgramData for easy cleanup
 $AgentBasePath = "$env:ProgramData\AutopilotMonitor"
@@ -105,89 +134,151 @@ $AgentBinPath = "$AgentBasePath\Agent"
 $AgentLogPath = "$AgentBasePath\Logs"
 $LogFile = "$AgentLogPath\bootstrap_agent.log"
 
-New-Item -Path $AgentBasePath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
-New-Item -Path $AgentBinPath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
-New-Item -Path $AgentLogPath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
-
+# Write-Host (NOT Write-Output): log calls happen inside functions that return
+# decision objects; Write-Output would leak every log line into those return values.
 function Write-Log {
     param([string]$Message)
     $timestamp = Get-Date -Format "yyyy-MM-dd HH:mm:ss"
     $logMessage = "[$timestamp] $Message"
-    Write-Output $logMessage
+    Write-Host $logMessage
     Add-Content -Path $LogFile -Value $logMessage -ErrorAction SilentlyContinue
 }
 
-# Positive OOBE check via the documented WinRT API
+# Raw OOBE state via the documented WinRT API
 # Windows.System.Profile.SystemSetupInfo.OutOfBoxExperienceState.
-# Returns $true ONLY when the OS reports InProgress. Any other state (Completed/NotStarted),
-# an older OS without the class (< Win10 1809 / 10.0.17763.0), or any error returns $false
+# Returns the state string (InProgress/Completed/NotStarted). An older OS without the
+# class (< Win10 1809 / 10.0.17763.0) or any error returns 'Unavailable'
 # -> SKIP-safe (we never relax a guard on uncertainty).
-# Docs: https://learn.microsoft.com/en-us/uwp/api/windows.system.profile.systemsetupinfo
-function Test-OobeInProgress {
+function Get-OobeState {
     try {
         $null = [Windows.System.Profile.SystemSetupInfo, Windows.System.Profile, ContentType = WindowsRuntime]
-        return ([Windows.System.Profile.SystemSetupInfo]::OutOfBoxExperienceState).ToString() -eq 'InProgress'
+        return ([Windows.System.Profile.SystemSetupInfo]::OutOfBoxExperienceState).ToString()
     }
     catch {
+        return 'Unavailable'
+    }
+}
+
+# Positive Windows 365 Cloud PC detection. Local WMI/SMBIOS on a Cloud PC only
+# reports the generic Manufacturer='Microsoft Corporation' / Model='Virtual Machine',
+# so identification uses two W365-specific markers instead. BOTH are required (AND):
+# each alone has look-alikes (e.g. W365 Boot physical clients, other managed-desktop
+# agents). Identity and marker states are logged for diagnosis; any error returns
+# $false (SKIP-safe).
+function Test-IsCloudPc {
+    try {
+        $cs = Get-CimInstance Win32_ComputerSystem -ErrorAction Stop
+        Write-Log "Computer identity: Manufacturer='$($cs.Manufacturer)' Model='$($cs.Model)'."
+    }
+    catch { Write-Log "INFO: computer identity query failed: $($_.Exception.Message)" }
+
+    try {
+        $regMarker = Test-Path 'HKLM:\SOFTWARE\Microsoft\Windows365'
+        $svcMarker = $null -ne (Get-Service -Name 'CloudManagedDesktopExtension' -ErrorAction SilentlyContinue)
+        Write-Log "Cloud PC markers: Windows365 registry key=$regMarker; CloudManagedDesktopExtension service=$svcMarker."
+        return ($regMarker -and $svcMarker)
+    }
+    catch {
+        Write-Log "INFO: Cloud PC marker check failed: $($_.Exception.Message)"
         return $false
     }
 }
 
-try {
-    Write-Log "===== Autopilot Monitor Bootstrap Started ====="
-    Write-Log "Bootstrap script version: v$ScriptVersion"
+# Collects real (non-special) user profiles as full paths (WMI Special flag + filesystem).
+# NOTE: callers MUST wrap the result in @(...). PowerShell unrolls function output, so an
+# empty result returns $null and a single path returns a scalar string, whose [0] indexer
+# would yield its first char ('C') instead of the full path.
+function Get-RealUserProfilePaths {
+    $excludePattern = '^(defaultuser\d*|Public|Default( User)?|All Users|WDAGUtilityAccount)$'
+    @(
+        try {
+            Get-CimInstance Win32_UserProfile -ErrorAction Stop |
+                Where-Object { -not $_.Special -and $_.LocalPath -like 'C:\Users\*' } |
+                ForEach-Object { $_.LocalPath }
+        } catch { Write-Log "INFO: WMI profile query failed, continuing with filesystem check." }
+
+        (Get-ChildItem 'C:\Users' -Directory -ErrorAction SilentlyContinue).FullName
+    ) | Where-Object { $_ -and (Split-Path $_ -Leaf) -notmatch $excludePattern } |
+        Select-Object -Unique
+}
+
+# Relax: the ONLY exemption to guards 2+3. Two triggers (see .DESCRIPTION):
+#   (1) OOBE InProgress          -> Windows Backup for Organizations restore
+#   (2) Cloud PC identity        -> Windows 365 (headless OOBE, profile at first connect)
+# Either trigger additionally requires: exactly one real profile AND that profile
+# created < $OobeProfileMaxAgeMinutes minutes ago. Any miss keeps the original skip.
+# Returns an object with: Active (bool), OobeState (string), IsCloudPc (bool).
+function Get-RelaxDecision {
+    param(
+        [string[]]$ProfilePaths,
+        [int]$OobeProfileMaxAgeMinutes
+    )
+    if ($null -eq $ProfilePaths) { $ProfilePaths = @() }
+
+    $relaxActive = $false
+    $oobeState = Get-OobeState
+    $oobeInProgress = ($oobeState -eq 'InProgress')
+    $isCloudPc = Test-IsCloudPc
+    if ($isCloudPc) {
+        Write-Log "Cloud PC detected (OOBE state: $oobeState)."
+    }
+
+    $relaxTrigger = $null
+    if ($oobeInProgress) {
+        $relaxTrigger = "OOBE InProgress (Windows Backup restore case)"
+    } elseif ($isCloudPc) {
+        $relaxTrigger = "Cloud PC (headless provisioning, OOBE state: $oobeState)"
+    }
+
+    if ($ProfilePaths.Count -eq 1 -and $relaxTrigger) {
+        try {
+            $created = (Get-Item -LiteralPath $ProfilePaths[0] -Force -ErrorAction Stop).CreationTimeUtc
+            $ageMin = ((Get-Date).ToUniversalTime() - $created).TotalMinutes
+            if ($ageMin -ge 0 -and $ageMin -lt $OobeProfileMaxAgeMinutes) {
+                $relaxActive = $true
+                Write-Log ("Relax active: {0} + single profile created {1:N1}min ago (< {2}min). Guards 2+3 relaxed." -f $relaxTrigger, $ageMin, $OobeProfileMaxAgeMinutes)
+            } else {
+                Write-Log ("Relax NOT applied: {0} + single profile '{1}', but profile created {2:N1}min ago (created {3:u}) is outside the [0, {4})min window." -f $relaxTrigger, (Split-Path $ProfilePaths[0] -Leaf), $ageMin, $created, $OobeProfileMaxAgeMinutes)
+            }
+        } catch { Write-Log "INFO: profile age check failed; relax not applied. $($_.Exception.Message)" }
+    } elseif ($ProfilePaths.Count -eq 1) {
+        Write-Log "Relax NOT applied: single profile present but neither OOBE InProgress nor Cloud PC (OOBE state: $oobeState)."
+    } elseif ($ProfilePaths.Count -gt 1) {
+        Write-Log ("Relax NOT applied: {0} real profiles present (relax requires exactly one). OOBE state: {1}; CloudPC: {2}." -f $ProfilePaths.Count, $oobeState, $isCloudPc)
+    }
+
+    return [pscustomobject]@{
+        Active    = $relaxActive
+        OobeState = $oobeState
+        IsCloudPc = $isCloudPc
+    }
+}
+
+# Runs pre-flight guards 1-5 and returns the install decision WITHOUT side effects
+# beyond logging. Returns an object with:
+#   Install (bool), ReasonCode (null | AlreadyDeployed | ProfilesFound |
+#   LastLoggedOnUser | UptimeExceeded | AgentPresent), RelaxActive (bool).
+function Get-BootstrapDecision {
+    param(
+        [int]$MaxBootstrapWindowHours,
+        [int]$OobeProfileMaxAgeMinutes,
+        [string]$AgentBinPath
+    )
 
     # Guard 1: Agent was already deployed on this device (registry marker survives self-destruct)
     $deployed = (Get-ItemProperty -Path 'HKLM:\SOFTWARE\AutopilotMonitor' -Name 'Deployed' -ErrorAction SilentlyContinue).Deployed
     if ($deployed) {
         Write-Log "SKIP: Agent was previously deployed at $deployed."
-        exit 0
+        return [pscustomobject]@{ Install = $false; ReasonCode = 'AlreadyDeployed'; RelaxActive = $false }
     }
 
-    # Collect real (non-special) user profiles as full paths (WMI Special flag + filesystem).
-    $excludePattern = '^(defaultuser\d*|Public|Default( User)?|All Users|WDAGUtilityAccount)$'
-
-    # NOTE: the outer @() is required. Without it the trailing Where/Select pipeline
-    # unwraps a single result back to a scalar string, so $profilePaths[0] would index
-    # into the string and return its first char ('C') instead of the full path.
-    $profilePaths = @(
-        @(
-            try {
-                Get-CimInstance Win32_UserProfile -ErrorAction Stop |
-                    Where-Object { -not $_.Special -and $_.LocalPath -like 'C:\Users\*' } |
-                    ForEach-Object { $_.LocalPath }
-            } catch { Write-Log "INFO: WMI profile query failed, continuing with filesystem check." }
-
-            (Get-ChildItem 'C:\Users' -Directory -ErrorAction SilentlyContinue).FullName
-        ) | Where-Object { $_ -and (Split-Path $_ -Leaf) -notmatch $excludePattern } |
-            Select-Object -Unique
-    )
-
-    # OOBE-relax: the ONLY exemption to guards 2+3. Matches exactly the Windows Backup for
-    # Organizations restore (one user signs in + restores DURING OOBE). All three must hold:
-    #   OOBE InProgress  AND  exactly one profile  AND  that profile created < N minutes ago.
-    $oobeRelax = $false
-    $oobeInProgress = Test-OobeInProgress
-    if ($profilePaths.Count -eq 1 -and $oobeInProgress) {
-        try {
-            $created = (Get-Item -LiteralPath $profilePaths[0] -Force -ErrorAction Stop).CreationTimeUtc
-            $ageMin = ((Get-Date).ToUniversalTime() - $created).TotalMinutes
-            if ($ageMin -ge 0 -and $ageMin -lt $OobeProfileMaxAgeMinutes) {
-                $oobeRelax = $true
-                Write-Log ("OOBE-relax active: OOBE InProgress + single profile created {0:N1}min ago (< {1}min). Guards 2+3 relaxed (Windows Backup restore case)." -f $ageMin, $OobeProfileMaxAgeMinutes)
-            } else {
-                Write-Log ("OOBE-relax NOT applied: OOBE InProgress + single profile '{0}', but profile created {1:N1}min ago (created {2:u}) is outside the [0, {3})min window." -f (Split-Path $profilePaths[0] -Leaf), $ageMin, $created, $OobeProfileMaxAgeMinutes)
-            }
-        } catch { Write-Log "INFO: profile age check failed; OOBE-relax not applied. $($_.Exception.Message)" }
-    } elseif ($profilePaths.Count -eq 1 -and -not $oobeInProgress) {
-        Write-Log "OOBE-relax NOT applied: single profile present but OutOfBoxExperienceState != InProgress (OOBE already finished / not detectable)."
-    } elseif ($profilePaths.Count -gt 1) {
-        Write-Log ("OOBE-relax NOT applied: {0} real profiles present (relax requires exactly one)." -f $profilePaths.Count)
-    }
+    # The @() wrap is required, see NOTE on Get-RealUserProfilePaths.
+    $profilePaths = @(Get-RealUserProfilePaths)
+    $relax = Get-RelaxDecision -ProfilePaths $profilePaths -OobeProfileMaxAgeMinutes $OobeProfileMaxAgeMinutes
 
     # Guard 2: No real user profile should exist yet (primary productive-device guard),
-    # unless the OOBE-relax signature holds.
-    if ($profilePaths.Count -gt 0 -and -not $oobeRelax) {
+    # unless the relax signature holds.
+    if ($profilePaths.Count -gt 0 -and -not $relax.Active) {
         $details = (($profilePaths | ForEach-Object {
             $leaf = Split-Path $_ -Leaf
             try {
@@ -195,26 +286,33 @@ try {
                 "{0} (created {1:u}, {2:N1}min ago)" -f $leaf, $c, ((Get-Date).ToUniversalTime() - $c).TotalMinutes
             } catch { "$leaf (age unknown)" }
         }) | Select-Object -First 3) -join '; '
-        Write-Log "SKIP: Real user profile(s) found ($details). OOBE InProgress=$oobeInProgress. Device appears productive."
-        exit 0
+        Write-Log "SKIP: Real user profile(s) found ($details). OOBE state: $($relax.OobeState). CloudPC: $($relax.IsCloudPc). Device appears productive."
+        return [pscustomobject]@{ Install = $false; ReasonCode = 'ProfilesFound'; RelaxActive = $false }
     }
 
     # Guard 3: LastLoggedOnUser - during Device ESP no real user has logged on yet
-    # (same OOBE-relax exemption: the restoring user signs in during OOBE).
+    # (same relax exemption: the restoring/first-connect user signs in legitimately).
     $lastLoggedOnUser = (Get-ItemProperty -Path 'HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Authentication\LogonUI' -Name 'LastLoggedOnUser' -EA SilentlyContinue).LastLoggedOnUser
-    if ($lastLoggedOnUser -and $lastLoggedOnUser -notmatch 'defaultuser\d*' -and -not $oobeRelax) {
+    if ($lastLoggedOnUser -and $lastLoggedOnUser -notmatch 'defaultuser\d*' -and -not $relax.Active) {
         Write-Log "SKIP: LastLoggedOnUser found ($lastLoggedOnUser). Device appears productive."
-        exit 0
+        return [pscustomobject]@{ Install = $false; ReasonCode = 'LastLoggedOnUser'; RelaxActive = $false }
     }
 
     # Guard 4: Bootstrap window check - device uptime must be within accepted OOBE window.
     # Sleep/standby does not reset uptime, only real boot/restart does.
+    # Exempted while the Cloud PC relax is active: a Cloud PC runs headless from
+    # provisioning until the user's first connect (often days), so boot uptime says
+    # nothing there -- the fresh-profile window of the relax is the time anchor.
     $lastBoot = (Get-CimInstance Win32_OperatingSystem).LastBootUpTime
     $uptimeHours = ((Get-Date) - $lastBoot).TotalHours
     Write-Log "Device uptime: $([int]$uptimeHours)h (last boot: $lastBoot)"
     if ($uptimeHours -gt $MaxBootstrapWindowHours) {
-        Write-Log "SKIP: Device uptime is $([int]$uptimeHours)h. OOBE state is older than accepted bootstrap window of ${MaxBootstrapWindowHours}h."
-        exit 0
+        if ($relax.Active -and $relax.IsCloudPc) {
+            Write-Log "Uptime $([int]$uptimeHours)h exceeds the ${MaxBootstrapWindowHours}h window, but the Cloud PC relax is active (fresh first-connect profile is the time anchor). Continuing."
+        } else {
+            Write-Log "SKIP: Device uptime is $([int]$uptimeHours)h. OOBE state is older than accepted bootstrap window of ${MaxBootstrapWindowHours}h."
+            return [pscustomobject]@{ Install = $false; ReasonCode = 'UptimeExceeded'; RelaxActive = $relax.Active }
+        }
     }
 
     # Guard 5: Is the agent already installed? (leftover from previous run)
@@ -222,148 +320,175 @@ try {
         $existingAgent = Get-ChildItem -Path $AgentBinPath -Filter "AutopilotMonitor.Agent.exe" -ErrorAction SilentlyContinue
         if ($existingAgent) {
             Write-Log "SKIP: Agent already installed at $($existingAgent.FullName)."
+            return [pscustomobject]@{ Install = $false; ReasonCode = 'AgentPresent'; RelaxActive = $relax.Active }
+        }
+    }
+
+    return [pscustomobject]@{ Install = $true; ReasonCode = $null; RelaxActive = $relax.Active }
+}
+
+# Main bootstrap flow. Only invoked when the script is run directly (see entry guard
+# at the bottom); never on dot-source.
+function Invoke-Bootstrap {
+    try {
+        New-Item -Path $AgentBasePath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
+        New-Item -Path $AgentBinPath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
+        New-Item -Path $AgentLogPath -ItemType Directory -Force -ErrorAction SilentlyContinue | Out-Null
+
+        Write-Log "===== Autopilot Monitor Bootstrap Started ====="
+        Write-Log "Bootstrap script version: v$ScriptVersion"
+
+        $decision = Get-BootstrapDecision -MaxBootstrapWindowHours $MaxBootstrapWindowHours -OobeProfileMaxAgeMinutes $OobeProfileMaxAgeMinutes -AgentBinPath $AgentBinPath
+        if (-not $decision.Install) {
             exit 0
         }
-    }
 
-    Write-Log "Pre-flight checks passed -- no prior deployment, no productive-device signals, within bootstrap window (oobeRelax=$oobeRelax)"
+        Write-Log "Pre-flight checks passed -- no prior deployment, no productive-device signals, within bootstrap window (relaxActive=$($decision.RelaxActive))"
 
-    # Download and extract agent binaries
-    $agentExePath = Join-Path $AgentBinPath "AutopilotMonitor.Agent.exe"
+        # Download and extract agent binaries
+        $agentExePath = Join-Path $AgentBinPath "AutopilotMonitor.Agent.exe"
 
-    if (Test-Path $agentExePath) {
-        Write-Log "Agent already installed at $agentExePath"
-    }
-    else {
-        Write-Log "Downloading agent from $AgentDownloadUrl..."
-
-        try {
-            # Derive manifest URL from the agent download URL (same blob container)
-            $versionJsonUrl = $AgentDownloadUrl -replace '[^/]+$', $VersionJsonName
-            $expectedSha256 = $null
-            $manifestAttempts = 3
-            $manifestDelays = @(1, 2)   # short backoff between attempts; do not block enrollment
-
-            for ($attempt = 1; $attempt -le $manifestAttempts; $attempt++) {
-                try {
-                    Write-Log "Fetching $VersionJsonName (attempt ${attempt}/${manifestAttempts}) from $versionJsonUrl for integrity verification..."
-                    $versionJsonResponse = Invoke-RestMethod -Uri $versionJsonUrl -UseBasicParsing -TimeoutSec 5 -ErrorAction Stop
-                    if ($versionJsonResponse.sha256) {
-                        $expectedSha256 = $versionJsonResponse.sha256.ToLowerInvariant()
-                        Write-Log "SHA-256 hash from manifest: $expectedSha256 (version: $($versionJsonResponse.version))"
-                    } else {
-                        Write-Log "WARNING: Manifest has no sha256 field - integrity cannot be verified."
-                    }
-                    break   # manifest reached (with or without hash); retrying would not help
-                }
-                catch {
-                    Write-Log "WARNING: Could not fetch integrity manifest (attempt $attempt): $($_.Exception.Message)"
-                    if ($attempt -lt $manifestAttempts) {
-                        Start-Sleep -Seconds $manifestDelays[$attempt - 1]
-                    }
-                }
-            }
-
-            $zipPath = Join-Path $env:TEMP "AutopilotMonitor-Agent.zip"
-            $maxDownloadAttempts = 3
-            $downloadAttempt = 0
-
-            do {
-                $downloadAttempt++
-                try {
-                    Write-Log "Download attempt ${downloadAttempt}/${maxDownloadAttempts}"
-                    $downloadStopwatch = [System.Diagnostics.Stopwatch]::StartNew()
-                    Invoke-WebRequest `
-                        -Uri $AgentDownloadUrl `
-                        -OutFile $zipPath `
-                        -UseBasicParsing `
-                        -TimeoutSec 30 `
-                        -ErrorAction Stop
-                    $downloadStopwatch.Stop()
-                    $downloadSeconds = [math]::Round($downloadStopwatch.Elapsed.TotalSeconds, 1)
-                    Write-Log "Downloaded agent to $zipPath (took ${downloadSeconds}s)"
-                    break
-                }
-                catch {
-                    if ($downloadAttempt -ge $maxDownloadAttempts) {
-                        throw
-                    }
-
-                    $retryDelaysInSeconds = @(2, 4, 8)
-                    $retryDelaySeconds = $retryDelaysInSeconds[$downloadAttempt - 1]
-                    Write-Log "Download failed (attempt $downloadAttempt): $($_.Exception.Message). Retrying in ${retryDelaySeconds}s..."
-                    Start-Sleep -Seconds $retryDelaySeconds
-                }
-            } while ($downloadAttempt -lt $maxDownloadAttempts)
-
-            # SHA-256 from the version manifest is the only accepted integrity proof.
-            # No hash -> refuse to install an unverified ZIP (SKIP-safe security posture).
-            if (-not $expectedSha256) {
-                throw "No SHA-256 hash available from manifest - refusing to install unverified agent ZIP."
-            }
-            $actualSha256 = (Get-FileHash $zipPath -Algorithm SHA256).Hash.ToLowerInvariant()
-            Write-Log "Validating SHA-256 hash against manifest"
-            if ($actualSha256 -ne $expectedSha256) {
-                throw "SHA-256 integrity check FAILED. Expected='$expectedSha256', Actual='$actualSha256'. Download may be tampered or corrupted."
-            }
-            Write-Log "SHA-256 integrity check passed"
-
-            Expand-Archive -Path $zipPath -DestinationPath $AgentBinPath -Force
-            Write-Log "Extracted agent to $AgentBinPath"
-
-            Remove-Item $zipPath -Force -ErrorAction SilentlyContinue
-            Write-Log "Cleaned up temporary files"
-
-            if (-not (Test-Path $agentExePath)) {
-                throw "Agent executable not found after extraction at $agentExePath"
-            }
-
-            Write-Log "Agent installation completed successfully"
+        if (Test-Path $agentExePath) {
+            Write-Log "Agent already installed at $agentExePath"
         }
-        catch {
-            Write-Log "ERROR downloading/extracting agent: $($_.Exception.Message)"
-            throw
+        else {
+            Write-Log "Downloading agent from $AgentDownloadUrl..."
+
+            try {
+                # Derive manifest URL from the agent download URL (same blob container)
+                $versionJsonUrl = $AgentDownloadUrl -replace '[^/]+$', $VersionJsonName
+                $expectedSha256 = $null
+                $manifestAttempts = 3
+                $manifestDelays = @(1, 2)   # short backoff between attempts; do not block enrollment
+
+                for ($attempt = 1; $attempt -le $manifestAttempts; $attempt++) {
+                    try {
+                        Write-Log "Fetching $VersionJsonName (attempt ${attempt}/${manifestAttempts}) from $versionJsonUrl for integrity verification..."
+                        $versionJsonResponse = Invoke-RestMethod -Uri $versionJsonUrl -UseBasicParsing -TimeoutSec 5 -ErrorAction Stop
+                        if ($versionJsonResponse.sha256) {
+                            $expectedSha256 = $versionJsonResponse.sha256.ToLowerInvariant()
+                            Write-Log "SHA-256 hash from manifest: $expectedSha256 (version: $($versionJsonResponse.version))"
+                        } else {
+                            Write-Log "WARNING: Manifest has no sha256 field - integrity cannot be verified."
+                        }
+                        break   # manifest reached (with or without hash); retrying would not help
+                    }
+                    catch {
+                        Write-Log "WARNING: Could not fetch integrity manifest (attempt $attempt): $($_.Exception.Message)"
+                        if ($attempt -lt $manifestAttempts) {
+                            Start-Sleep -Seconds $manifestDelays[$attempt - 1]
+                        }
+                    }
+                }
+
+                $zipPath = Join-Path $env:TEMP "AutopilotMonitor-Agent.zip"
+                $maxDownloadAttempts = 3
+                $downloadAttempt = 0
+
+                do {
+                    $downloadAttempt++
+                    try {
+                        Write-Log "Download attempt ${downloadAttempt}/${maxDownloadAttempts}"
+                        $downloadStopwatch = [System.Diagnostics.Stopwatch]::StartNew()
+                        Invoke-WebRequest `
+                            -Uri $AgentDownloadUrl `
+                            -OutFile $zipPath `
+                            -UseBasicParsing `
+                            -TimeoutSec 30 `
+                            -ErrorAction Stop
+                        $downloadStopwatch.Stop()
+                        $downloadSeconds = [math]::Round($downloadStopwatch.Elapsed.TotalSeconds, 1)
+                        Write-Log "Downloaded agent to $zipPath (took ${downloadSeconds}s)"
+                        break
+                    }
+                    catch {
+                        if ($downloadAttempt -ge $maxDownloadAttempts) {
+                            throw
+                        }
+
+                        $retryDelaysInSeconds = @(2, 4, 8)
+                        $retryDelaySeconds = $retryDelaysInSeconds[$downloadAttempt - 1]
+                        Write-Log "Download failed (attempt $downloadAttempt): $($_.Exception.Message). Retrying in ${retryDelaySeconds}s..."
+                        Start-Sleep -Seconds $retryDelaySeconds
+                    }
+                } while ($downloadAttempt -lt $maxDownloadAttempts)
+
+                # SHA-256 from the version manifest is the only accepted integrity proof.
+                # No hash -> refuse to install an unverified ZIP (SKIP-safe security posture).
+                if (-not $expectedSha256) {
+                    throw "No SHA-256 hash available from manifest - refusing to install unverified agent ZIP."
+                }
+                $actualSha256 = (Get-FileHash $zipPath -Algorithm SHA256).Hash.ToLowerInvariant()
+                Write-Log "Validating SHA-256 hash against manifest"
+                if ($actualSha256 -ne $expectedSha256) {
+                    throw "SHA-256 integrity check FAILED. Expected='$expectedSha256', Actual='$actualSha256'. Download may be tampered or corrupted."
+                }
+                Write-Log "SHA-256 integrity check passed"
+
+                Expand-Archive -Path $zipPath -DestinationPath $AgentBinPath -Force
+                Write-Log "Extracted agent to $AgentBinPath"
+
+                Remove-Item $zipPath -Force -ErrorAction SilentlyContinue
+                Write-Log "Cleaned up temporary files"
+
+                if (-not (Test-Path $agentExePath)) {
+                    throw "Agent executable not found after extraction at $agentExePath"
+                }
+
+                Write-Log "Agent installation completed successfully"
+            }
+            catch {
+                Write-Log "ERROR downloading/extracting agent: $($_.Exception.Message)"
+                throw
+            }
         }
-    }
 
-    Write-Log "Calling agent install mode (--install)..."
-    & $agentExePath --install
-    $installExitCode = $LASTEXITCODE
-    if ($installExitCode -ne 0) {
-        throw "Agent install failed with exit code $installExitCode"
-    }
-    Write-Log "Agent install mode completed successfully"
+        Write-Log "Calling agent install mode (--install)..."
+        & $agentExePath --install
+        $installExitCode = $LASTEXITCODE
+        if ($installExitCode -ne 0) {
+            throw "Agent install failed with exit code $installExitCode"
+        }
+        Write-Log "Agent install mode completed successfully"
 
-    $runtimeProcessName = 'AutopilotMonitor.Agent'
-    $verifyTimeoutSec = 10
-    $verifyDeadline = (Get-Date).AddSeconds($verifyTimeoutSec)
-    $runtimeProc = $null
-    while ((Get-Date) -lt $verifyDeadline) {
-        $runtimeProc = Get-Process -Name $runtimeProcessName -ErrorAction SilentlyContinue | Select-Object -First 1
-        if ($null -ne $runtimeProc) { break }
-        Start-Sleep -Milliseconds 500
-    }
-    if ($null -ne $runtimeProc) {
-        $startedUtc = 'unavailable'
-        try { $startedUtc = $runtimeProc.StartTime.ToUniversalTime().ToString('o') } catch { }
-        Write-Log ("Runtime process verified: name={0}.exe pid={1} startedUtc={2}" -f $runtimeProcessName, $runtimeProc.Id, $startedUtc)
-    } else {
-        Write-Log ("WARNING: Runtime process verification FAILED. Agent --install reported success but no '{0}.exe' process appeared within {1}s. Likely silent block (AV/EDR, AppLocker/WDAC) of the runtime launch. Agent should still come up at next boot via the BootTrigger task. Check Event Viewer > Microsoft > Windows > TaskScheduler/Operational and AV/EDR logs for '{0}.exe'." -f $runtimeProcessName, $verifyTimeoutSec)
-    }
+        $runtimeProcessName = 'AutopilotMonitor.Agent'
+        $verifyTimeoutSec = 10
+        $verifyDeadline = (Get-Date).AddSeconds($verifyTimeoutSec)
+        $runtimeProc = $null
+        while ((Get-Date) -lt $verifyDeadline) {
+            $runtimeProc = Get-Process -Name $runtimeProcessName -ErrorAction SilentlyContinue | Select-Object -First 1
+            if ($null -ne $runtimeProc) { break }
+            Start-Sleep -Milliseconds 500
+        }
+        if ($null -ne $runtimeProc) {
+            $startedUtc = 'unavailable'
+            try { $startedUtc = $runtimeProc.StartTime.ToUniversalTime().ToString('o') } catch { }
+            Write-Log ("Runtime process verified: name={0}.exe pid={1} startedUtc={2}" -f $runtimeProcessName, $runtimeProc.Id, $startedUtc)
+        } else {
+            Write-Log ("WARNING: Runtime process verification FAILED. Agent --install reported success but no '{0}.exe' process appeared within {1}s. Likely silent block (AV/EDR, AppLocker/WDAC) of the runtime launch. Agent should still come up at next boot via the BootTrigger task. Check Event Viewer > Microsoft > Windows > TaskScheduler/Operational and AV/EDR logs for '{0}.exe'." -f $runtimeProcessName, $verifyTimeoutSec)
+        }
 
-    Write-Log "===== Bootstrap Completed Successfully ====="
+        Write-Log "===== Bootstrap Completed Successfully ====="
 
-    exit 0
+        exit 0
+    }
+    catch {
+        Write-Log "===== Bootstrap FAILED ====="
+        Write-Log "ERROR: $($_.Exception.Message)"
+        Write-Log "Stack trace: $($_.ScriptStackTrace)"
+        Write-Log "Please check log file: $LogFile"
+
+        $errMsg = "AutopilotMonitor bootstrap failed: $($_.Exception.Message)"
+        if ($errMsg.Length -gt 2048) { $errMsg = $errMsg.Substring(0, 1045) + '...' }
+        [Console]::Error.WriteLine($errMsg)
+
+        exit 1
+    }
 }
-catch {
-    Write-Log "===== Bootstrap FAILED ====="
-    Write-Log "ERROR: $($_.Exception.Message)"
-    Write-Log "Stack trace: $($_.ScriptStackTrace)"
-    Write-Log "Please check log file: $LogFile"
 
-    $errMsg = "AutopilotMonitor bootstrap failed: $($_.Exception.Message)"
-    if ($errMsg.Length -gt 2048) { $errMsg = $errMsg.Substring(0, 1045) + '...' }
-    [Console]::Error.WriteLine($errMsg)
-
-    exit 1
+# Entry guard: execute only when run as a script (IME, console). Dot-sourcing
+# (". .\Install-AutopilotMonitor.ps1", as the Pester suite does) only loads
+# the functions above and MUST stay side-effect free.
+if ($MyInvocation.InvocationName -ne '.') {
+    Invoke-Bootstrap
 }

scripts/Bootstrap/Test-ShouldBootstrapAgent.ps1

@@ -1,7 +1,9 @@
 # Autopilot Monitor Bootstrap Guard Test
 # Dry-run only: does not exit, does not install, only reports what WOULD happen.
-# Mirrors Install-AutopilotMonitor.ps1 incl. the OOBE-relax exemption for the
-# "Windows Backup for Organizations" restore case (guards 2+3).
+# Mirrors Install-AutopilotMonitor.ps1 (v2.3) incl. the relax exemption for
+# guards 2+3, triggered by "Windows Backup for Organizations" (OOBE InProgress)
+# or a positively identified Windows 365 Cloud PC, and the guard 4 (uptime)
+# exemption while the Cloud PC relax is active.
 
 param(
     [string]$AgentBinPath = 'C:\ProgramData\AutopilotMonitor\Agent',
@@ -54,6 +56,12 @@ function Test-OobeInProgress {
     }
 }
 
+# Positive Windows 365 Cloud PC detection: two local markers, BOTH required (AND),
+# because each alone has look-alikes. Mirrors Test-IsCloudPc in the bootstrapper.
+$cloudPcRegMarker = Test-Path 'HKLM:\SOFTWARE\Microsoft\Windows365'
+$cloudPcSvcMarker = $null -ne (Get-Service -Name 'CloudManagedDesktopExtension' -ErrorAction SilentlyContinue)
+$isCloudPc = ($cloudPcRegMarker -and $cloudPcSvcMarker)
+
 $AgentExePath = Join-Path $AgentBinPath 'AutopilotMonitor.Agent.exe'
 $MaxBootstrapWindowHours = 12
 
@@ -104,8 +112,9 @@ if ($wmiProfileQueryFailed) {
 
 $profileNames = $profilePaths | ForEach-Object { Split-Path $_ -Leaf }
 
-# OOBE-relax: the ONLY exemption to guards 2+3. Matches exactly the Windows Backup for
-# Organizations restore: OOBE InProgress AND exactly one profile AND that profile fresh.
+# Relax: the ONLY exemption to guards 2+3. Two triggers -- OOBE InProgress (Windows
+# Backup for Organizations restore) or Cloud PC identity (headless provisioning,
+# profile at first connect). Either additionally requires exactly one fresh profile.
 $oobeInProgress = Test-OobeInProgress
 $profileAgeMin = $null
 $oobeRelax = $false
@@ -114,12 +123,13 @@ if ($profilePaths.Count -eq 1) {
         $created = (Get-Item -LiteralPath $profilePaths[0] -Force -ErrorAction Stop).CreationTimeUtc
         $profileAgeMin = [math]::Round(((Get-Date).ToUniversalTime() - $created).TotalMinutes, 1)
     } catch { }
-    if ($oobeInProgress -and $null -ne $profileAgeMin -and $profileAgeMin -ge 0 -and $profileAgeMin -lt $OobeProfileMaxAgeMinutes) {
+    if (($oobeInProgress -or $isCloudPc) -and $null -ne $profileAgeMin -and $profileAgeMin -ge 0 -and $profileAgeMin -lt $OobeProfileMaxAgeMinutes) {
         $oobeRelax = $true
     }
 }
 
-Write-Step -Status 'INFO' -Message "OOBE: OutOfBoxExperienceState InProgress = $oobeInProgress ; profileCount = $($profilePaths.Count) ; singleProfileAgeMin = $profileAgeMin ; OOBE-relax = $oobeRelax"
+Write-Step -Status 'INFO' -Message "Cloud PC markers: Windows365 registry key = $cloudPcRegMarker ; CloudManagedDesktopExtension service = $cloudPcSvcMarker ; IsCloudPc = $isCloudPc"
+Write-Step -Status 'INFO' -Message "OOBE: OutOfBoxExperienceState InProgress = $oobeInProgress ; profileCount = $($profilePaths.Count) ; singleProfileAgeMin = $profileAgeMin ; relax = $oobeRelax"
 
 # Guard 2: No real user profile should exist yet (unless OOBE-relax holds)
 if ($profilePaths.Count -gt 0 -and -not $oobeRelax) {
@@ -129,7 +139,8 @@ if ($profilePaths.Count -gt 0 -and -not $oobeRelax) {
     $reasons.Add("Real user profile(s) found: $names")
 } elseif ($profilePaths.Count -gt 0 -and $oobeRelax) {
     $names = ($profileNames | Select-Object -First 3) -join ', '
-    Write-Step -Status 'PASS' -Message "Guard 2: Profile ($names) present but OOBE-relax active (single fresh profile in OOBE) -- not blocking."
+    $trigger = if ($oobeInProgress) { 'OOBE InProgress' } else { 'Cloud PC' }
+    Write-Step -Status 'PASS' -Message "Guard 2: Profile ($names) present but relax active ($trigger + single fresh profile) -- not blocking."
 } else {
     Write-Step -Status 'PASS' -Message 'Guard 2: No real user profiles found (combined WMI/filesystem view).'
 }
@@ -141,7 +152,7 @@ if ($lastLoggedOnUser -and $lastLoggedOnUser -notmatch 'defaultuser\d*' -and -no
     $wouldInstall = $false
     $reasons.Add("LastLoggedOnUser found: $lastLoggedOnUser")
 } elseif ($lastLoggedOnUser -and $lastLoggedOnUser -notmatch 'defaultuser\d*' -and $oobeRelax) {
-    Write-Step -Status 'PASS' -Message "Guard 3: LastLoggedOnUser ($lastLoggedOnUser) present but OOBE-relax active -- not blocking."
+    Write-Step -Status 'PASS' -Message "Guard 3: LastLoggedOnUser ($lastLoggedOnUser) present but relax active -- not blocking."
 } else {
     Write-Step -Status 'PASS' -Message "Guard 3: No real LastLoggedOnUser (value: '$lastLoggedOnUser')."
 }
@@ -155,9 +166,15 @@ try {
     $uptimeHours = ((Get-Date) - $lastBoot).TotalHours
 
     if ($uptimeHours -gt $MaxBootstrapWindowHours) {
-        Write-Step -Status 'SKIP' -Message "Guard 4: Device uptime is $([int]$uptimeHours)h. Older than accepted bootstrap window of ${MaxBootstrapWindowHours}h."
-        $wouldInstall = $false
-        $reasons.Add("Uptime exceeds bootstrap window: $([int]$uptimeHours)h > ${MaxBootstrapWindowHours}h")
+        if ($oobeRelax -and $isCloudPc) {
+            # Cloud PCs run headless for days before the user's first connect; the
+            # fresh-profile window is the time anchor there, not boot uptime.
+            Write-Step -Status 'PASS' -Message "Guard 4: Device uptime is $([int]$uptimeHours)h (over ${MaxBootstrapWindowHours}h), but the Cloud PC relax is active -- not blocking."
+        } else {
+            Write-Step -Status 'SKIP' -Message "Guard 4: Device uptime is $([int]$uptimeHours)h. Older than accepted bootstrap window of ${MaxBootstrapWindowHours}h."
+            $wouldInstall = $false
+            $reasons.Add("Uptime exceeds bootstrap window: $([int]$uptimeHours)h > ${MaxBootstrapWindowHours}h")
+        }
     } else {
         Write-Step -Status 'PASS' -Message "Guard 4: Device uptime is $([int]$uptimeHours)h and within bootstrap window of ${MaxBootstrapWindowHours}h."
     }
@@ -202,9 +219,12 @@ Write-Host "Deployed marker           : $deployed"
 Write-Host "DetectedProfileNames      : $($profileNames -join ', ')"
 Write-Host "ProfileCount              : $($profilePaths.Count)"
 Write-Host "OobeInProgress (WinRT)    : $oobeInProgress"
+Write-Host "CloudPcRegistryMarker     : $cloudPcRegMarker"
+Write-Host "CloudPcServiceMarker      : $cloudPcSvcMarker"
+Write-Host "IsCloudPc                 : $isCloudPc"
 Write-Host "SingleProfileAgeMinutes   : $profileAgeMin"
 Write-Host "OobeProfileMaxAgeMinutes  : $OobeProfileMaxAgeMinutes"
-Write-Host "OobeRelaxActive           : $oobeRelax"
+Write-Host "RelaxActive               : $oobeRelax"
 Write-Host "LastLoggedOnUser          : $lastLoggedOnUser"
 Write-Host "MaxBootstrapWindowHours   : $MaxBootstrapWindowHours"
 Write-Host "LastBootUpTime            : $lastBoot"

src/Agent/AutopilotMonitor.Agent.V2.Core.Tests/Logging/AgentLoggerTests.cs

@@ -94,6 +94,21 @@ namespace AutopilotMonitor.Agent.V2.Core.Tests.Logging
             Assert.Equal("older-002\n", File.ReadAllText(Path.Combine(tmp.Path, TodayBaseName + "_002.log")));
         }
 
+        [Fact]
+        public void Log_transliterates_typographic_punctuation_to_ascii()
+        {
+            using var tmp = new TempDirectory();
+            var logger = new AgentLogger(tmp.Path, AgentLogLevel.Info, maxFileSizeBytes: 1024L * 1024);
+
+            // The log file is UTF-8 without BOM; cp1252-guessing viewers (Outlook web
+            // preview) render em dashes as mojibake. Our own punctuation must come out
+            // as plain ASCII, while genuine non-ASCII payload stays untouched.
+            logger.Info("check failed \u2014 range 1\u20132 \u2192 done\u2026 \u2018a\u2019 \u201Cb\u201D nb\u00A0sp f\u00FCr M\u00FCller");
+
+            var text = File.ReadAllText(Path.Combine(tmp.Path, TodayBaseName + ".log"));
+            Assert.Contains("check failed - range 1-2 -> done... 'a' \"b\" nb sp f\u00FCr M\u00FCller", text);
+        }
+
         [Fact]
         public void Log_swallows_io_errors_silently()
         {

src/Agent/AutopilotMonitor.Agent.V2.Core.Tests/Monitoring/Gather/GatherRuleCollectorGuardTests.cs

@@ -308,9 +308,52 @@ namespace AutopilotMonitor.Agent.V2.Core.Tests.Monitoring.Gather
         [InlineData("SELECT * FROM Win32_UserAccount")]
         [InlineData("SELECT * FROM Win32_ShadowCopy")]
         [InlineData("SELECT * FROM Win32_OperatingSystemExtra")]          // prefix spoofing
+        [InlineData("SELECT Name FROM Win32_Process")]                    // projection of a disallowed class
+        [InlineData("SELECT BatteryStatus FROM Win32_BatteryX")]          // class spoofing via projection
         public void WmiCollector_OffAllowlistQuery_IsBlockedBeforeTheQueryRuns(string query)
             => AssertBlocked(new WmiCollector().Execute(Rule("wmi", query), Context()), query);
 
+        // -------------------------------------------------------------------
+        // WMI query matcher: "SELECT <* | property-list> FROM <allowed class>"
+        //
+        // Guard-level (no WMI touched): a property projection of an allowed class
+        // exposes a strict subset of the already-allowed SELECT *, so it is admitted.
+        // -------------------------------------------------------------------
+
+        [Theory]
+        [InlineData("SELECT * FROM Win32_Battery")]                                       // legacy form unchanged
+        [InlineData("SELECT * FROM Win32_BIOS WHERE PrimaryBIOS = TRUE")]                 // legacy + WHERE
+        [InlineData("SELECT BatteryStatus FROM Win32_Battery")]
+        [InlineData("SELECT BatteryStatus, EstimatedChargeRemaining FROM Win32_Battery")]
+        [InlineData("select batterystatus from win32_battery")]                           // case-insensitive
+        [InlineData("SELECT\tBatteryStatus\tFROM\tWin32_Battery")]                        // tabs as separators
+        [InlineData("  SELECT BatteryStatus FROM Win32_Battery  ")]                       // surrounding whitespace
+        [InlineData("SELECT BatteryStatus,EstimatedChargeRemaining FROM Win32_Battery")]  // no space after comma
+        [InlineData("SELECT Status FROM Win32_Battery WHERE BatteryStatus = 1")]          // projection + WHERE
+        public void WmiGuard_ProjectionOfAllowedClass_IsAllowed(string query)
+            => Assert.True(GatherRuleGuards.IsWmiQueryAllowed(query));
+
+        [Theory]
+        [InlineData("SELECT Name FROM Win32_Process")]                    // class not on the allowlist
+        [InlineData("SELECT BatteryStatus FROM Win32_BatteryX")]          // class spoofing
+        [InlineData("SELECT * FROM Win32_BatteryX")]
+        [InlineData("SELECT *, Name FROM Win32_BIOS")]                    // star mixed into a property list
+        [InlineData("SELECT FROM Win32_BIOS")]                            // empty property list
+        [InlineData("SELECT BatteryStatus FROM")]                         // missing class
+        [InlineData("SELECT BatteryStatus FROM Win32_Battery;DROP")]      // non-whitespace boundary after class
+        [InlineData("SELECT BatteryStatus, FROM Win32_Battery")]          // trailing comma
+        [InlineData("")]
+        public void WmiGuard_MalformedOrDisallowedQuery_IsBlocked(string query)
+            => Assert.False(GatherRuleGuards.IsWmiQueryAllowed(query));
+
+        [Fact]
+        public void WmiGuard_NullQuery_IsBlocked()
+            => Assert.False(GatherRuleGuards.IsWmiQueryAllowed(null));
+
+        [Fact]
+        public void WmiGuard_UnrestrictedMode_AllowsAnyQuery()
+            => Assert.True(GatherRuleGuards.IsWmiQueryAllowed("SELECT Name FROM Win32_Process", unrestrictedMode: true));
+
         [Fact]
         public void JsonCollector_UserProfilePath_IsBlockedBeforeTheFileIsOpened()
         {

src/Agent/AutopilotMonitor.Agent.V2.Core.Tests/Program/InstallModeMergeTests.cs

@@ -132,10 +132,12 @@ namespace AutopilotMonitor.Agent.V2.Core.Tests.Program
         }
 
         [Fact]
-        public void First_install_without_wait_arg_uses_agent_default_of_600()
+        public void First_install_without_wait_arg_persists_null_so_runtime_default_applies()
         {
             // Generic bootstrap path: PS1 calls `--install` without --tenant-id-wait,
-            // and there is no persisted config yet. Agent owns the default (600 s).
+            // and there is no persisted config yet. "Not configured" is persisted as null
+            // so BuildAgentConfiguration applies the agent-side default (600 s) — freezing
+            // a number into the file would decouple deployed devices from default changes.
             var merged = AutopilotMonitor.Agent.V2.Program.MergeBootstrapConfig(
                 existing: null,
                 bootstrapTokenArg: null!, bootstrapTokenGiven: false,
@@ -144,20 +146,42 @@ namespace AutopilotMonitor.Agent.V2.Core.Tests.Program
 
             Assert.Null(merged.BootstrapToken);
             Assert.Null(merged.TenantId);
-            Assert.Equal(600, merged.TenantIdWaitSeconds);
+            Assert.Null(merged.TenantIdWaitSeconds);
         }
 
         [Fact]
-        public void Existing_config_without_wait_field_treated_as_zero()
+        public void Legacy_config_without_wait_field_stays_null_on_plain_reinstall()
         {
-            // BackCompat: pre-fix bootstrap-config.json has no TenantIdWaitSeconds field;
-            // Newtonsoft deserialises that to 0. New install with --tenant-id-wait 600
-            // overrides cleanly; no extra migration logic required.
+            // BackCompat: pre-feature bootstrap-config.json has no TenantIdWaitSeconds
+            // property → nullable deserialisation yields null. A reinstall that does not
+            // pass --tenant-id-wait must NOT turn that into 0 (which would silently
+            // disable the wait); null carries through and the runtime default applies.
             var existing = new BootstrapConfigFile
             {
                 BootstrapToken = "tok-legacy",
                 TenantId = "tenant-legacy",
-                TenantIdWaitSeconds = 0,
+                TenantIdWaitSeconds = null,
+            };
+
+            var merged = AutopilotMonitor.Agent.V2.Program.MergeBootstrapConfig(
+                existing,
+                bootstrapTokenArg: null!, bootstrapTokenGiven: false,
+                tenantIdArg: null!, tenantIdGiven: false,
+                tenantIdWaitSeconds: 0, tenantIdWaitGiven: false);
+
+            Assert.Equal("tok-legacy", merged.BootstrapToken);
+            Assert.Equal("tenant-legacy", merged.TenantId);
+            Assert.Null(merged.TenantIdWaitSeconds);
+        }
+
+        [Fact]
+        public void Legacy_config_without_wait_field_takes_explicit_wait_arg()
+        {
+            var existing = new BootstrapConfigFile
+            {
+                BootstrapToken = "tok-legacy",
+                TenantId = "tenant-legacy",
+                TenantIdWaitSeconds = null,
             };
 
             var merged = AutopilotMonitor.Agent.V2.Program.MergeBootstrapConfig(
@@ -170,5 +194,26 @@ namespace AutopilotMonitor.Agent.V2.Core.Tests.Program
             Assert.Equal("tenant-legacy", merged.TenantId);
             Assert.Equal(600, merged.TenantIdWaitSeconds);
         }
+
+        [Fact]
+        public void Explicit_zero_opt_out_survives_redeploy_without_wait_arg()
+        {
+            // An admin's deliberate `--tenant-id-wait 0` must not be resurrected to a
+            // wait by a later redeploy that omits the flag.
+            var existing = new BootstrapConfigFile
+            {
+                BootstrapToken = "tok-abc",
+                TenantId = "tenant-xyz",
+                TenantIdWaitSeconds = 0,
+            };
+
+            var merged = AutopilotMonitor.Agent.V2.Program.MergeBootstrapConfig(
+                existing,
+                bootstrapTokenArg: null!, bootstrapTokenGiven: false,
+                tenantIdArg: null!, tenantIdGiven: false,
+                tenantIdWaitSeconds: 0, tenantIdWaitGiven: false);
+
+            Assert.Equal(0, merged.TenantIdWaitSeconds);
+        }
     }
 }

src/Agent/AutopilotMonitor.Agent.V2.Core.Tests/Program/TenantIdWaitResolutionTests.cs

@@ -0,0 +1,90 @@
+using AutopilotMonitor.Agent.V2;
+using AutopilotMonitor.Agent.V2.Core.Configuration;
+using Xunit;
+
+namespace AutopilotMonitor.Agent.V2.Core.Tests.Program
+{
+    /// <summary>
+    /// Pins the TenantIdWaitSeconds resolution order in
+    /// <see cref="AutopilotMonitor.Agent.V2.Program.BuildAgentConfiguration"/>:
+    /// CLI arg &gt; persisted bootstrap-config.json value &gt; agent default (600 s).
+    /// The regression this guards: a legacy bootstrap-config.json written before the
+    /// TenantIdWaitSeconds property existed deserialises to null — that must fall
+    /// through to the 600 s default, NOT be treated as an explicit 0 opt-out
+    /// (which silently disabled the wait and broke Hybrid-AAD-join devices whose
+    /// registry TenantId lands minutes after first boot).
+    /// </summary>
+    public sealed class TenantIdWaitResolutionTests
+    {
+        private static AgentConfiguration Build(BootstrapConfigFile bootstrapConfig, params string[] args) =>
+            AutopilotMonitor.Agent.V2.Program.BuildAgentConfiguration(
+                args: args,
+                tenantId: "tenant-t",
+                sessionId: "session-s",
+                bootstrapConfig: bootstrapConfig,
+                awaitConfig: null);
+
+        [Fact]
+        public void No_cli_and_no_bootstrap_config_uses_default_600()
+        {
+            var cfg = Build(bootstrapConfig: null);
+
+            Assert.Equal(600, cfg.TenantIdWaitSeconds);
+        }
+
+        [Fact]
+        public void Legacy_bootstrap_config_with_null_wait_uses_default_600()
+        {
+            var legacy = new BootstrapConfigFile
+            {
+                BootstrapToken = "tok-legacy",
+                TenantId = "tenant-legacy",
+                TenantIdWaitSeconds = null,
+            };
+
+            var cfg = Build(legacy);
+
+            Assert.Equal(600, cfg.TenantIdWaitSeconds);
+        }
+
+        [Fact]
+        public void Explicit_zero_in_bootstrap_config_disables_wait()
+        {
+            var optedOut = new BootstrapConfigFile { TenantIdWaitSeconds = 0 };
+
+            var cfg = Build(optedOut);
+
+            Assert.Equal(0, cfg.TenantIdWaitSeconds);
+        }
+
+        [Fact]
+        public void Nonzero_bootstrap_config_value_wins_over_default()
+        {
+            var configured = new BootstrapConfigFile { TenantIdWaitSeconds = 1800 };
+
+            var cfg = Build(configured);
+
+            Assert.Equal(1800, cfg.TenantIdWaitSeconds);
+        }
+
+        [Fact]
+        public void Cli_arg_wins_over_bootstrap_config_value()
+        {
+            var configured = new BootstrapConfigFile { TenantIdWaitSeconds = 1800 };
+
+            var cfg = Build(configured, "--tenant-id-wait", "300");
+
+            Assert.Equal(300, cfg.TenantIdWaitSeconds);
+        }
+
+        [Fact]
+        public void Cli_zero_wins_over_bootstrap_config_value()
+        {
+            var configured = new BootstrapConfigFile { TenantIdWaitSeconds = 1800 };
+
+            var cfg = Build(configured, "--tenant-id-wait", "0");
+
+            Assert.Equal(0, cfg.TenantIdWaitSeconds);
+        }
+    }
+}

src/Agent/AutopilotMonitor.Agent.V2.Core.Tests/Security/AuthFailureTrackerTests.cs

@@ -260,6 +260,60 @@ namespace AutopilotMonitor.Agent.V2.Core.Tests.Security
             Assert.Contains("agent/config", fakeDistress.Calls[0].Message);
         }
 
+        [Fact]
+        public void RecordFailure_endpoint_unavailable_suppresses_distress_but_still_counts()
+        {
+            var clock = new FakeClock(new DateTime(2026, 4, 21, 10, 0, 0, DateTimeKind.Utc));
+            var fakeDistress = new FakeDistressReporter();
+            var tracker = new AuthFailureTracker(maxFailures: 2, timeoutMinutes: 0, clock, NewLogger(), fakeDistress);
+
+            int events = 0;
+            tracker.ThresholdExceeded += (_, _) => events++;
+
+            // Platform-403s from a stopped endpoint: a distress report cannot reach it and the
+            // DeviceNotRegistered classification would be wrong — nothing must be dispatched.
+            tracker.RecordFailure(403, "agent/config", endpointUnavailable: true);
+            tracker.RecordFailure(403, "agent/register-session", endpointUnavailable: true);
+
+            Assert.Empty(fakeDistress.Calls);
+            // The shutdown thresholds still apply — a dead endpoint must not retry forever.
+            Assert.Equal(2, tracker.ConsecutiveFailures);
+            Assert.Equal(1, events);
+        }
+
+        [Fact]
+        public void RecordFailure_endpoint_unavailable_does_not_consume_streak_report_slot()
+        {
+            var clock = new FakeClock(new DateTime(2026, 4, 21, 10, 0, 0, DateTimeKind.Utc));
+            var fakeDistress = new FakeDistressReporter();
+            var tracker = new AuthFailureTracker(maxFailures: 50, timeoutMinutes: 0, clock, NewLogger(), fakeDistress);
+
+            // The streak opens with a platform-403 (no dispatch), then a live backend answers
+            // with a genuine 403 — that one must still produce the streak's single report.
+            tracker.RecordFailure(403, "agent/config", endpointUnavailable: true);
+            tracker.RecordFailure(403, "agent/config");
+            tracker.RecordFailure(403, "agent/config");
+
+            Assert.Single(fakeDistress.Calls);
+            Assert.Equal(DistressErrorType.DeviceNotRegistered, fakeDistress.Calls[0].ErrorType);
+        }
+
+        [Fact]
+        public void RecordSuccess_rearms_distress_dispatch_for_the_next_streak()
+        {
+            var clock = new FakeClock(new DateTime(2026, 4, 21, 10, 0, 0, DateTimeKind.Utc));
+            var fakeDistress = new FakeDistressReporter();
+            var tracker = new AuthFailureTracker(maxFailures: 50, timeoutMinutes: 0, clock, NewLogger(), fakeDistress);
+
+            tracker.RecordFailure(401, "agent/config");
+            tracker.RecordSuccess();
+            tracker.RecordFailure(403, "agent/config");
+
+            Assert.Equal(2, fakeDistress.Calls.Count);
+            Assert.Equal(DistressErrorType.AuthCertificateRejected, fakeDistress.Calls[0].ErrorType);
+            Assert.Equal(DistressErrorType.DeviceNotRegistered, fakeDistress.Calls[1].ErrorType);
+        }
+
         [Fact]
         public void RecordFailure_without_distress_reporter_is_silent_but_still_tracks()
         {

src/Agent/AutopilotMonitor.Agent.V2.Core.Tests/Transport/BackendApiClientAuthClassificationTests.cs

@@ -0,0 +1,110 @@
+#nullable enable
+using System;
+using System.Net;
+using System.Net.Http;
+using System.Text;
+using System.Threading;
+using System.Threading.Tasks;
+using AutopilotMonitor.Agent.V2.Core.Monitoring.Transport;
+using Xunit;
+
+namespace AutopilotMonitor.Agent.V2.Core.Tests.Transport
+{
+    /// <summary>
+    /// 401/403 classification in <see cref="BackendApiClient"/>: a backend JSON 403 is a real
+    /// device-authorization verdict, while an HTML body means the response came from the Azure
+    /// platform / an edge proxy (stopped or retired app) and must NOT be reported as
+    /// "device is not authorized". Field case sits-d.cloud 2026-08-09: a pre-cutover agent hit
+    /// the stopped legacy Function App and support chased a non-existent authorization problem.
+    /// </summary>
+    public sealed class BackendApiClientAuthClassificationTests
+    {
+        private sealed class StubHandler : HttpMessageHandler
+        {
+            private readonly Func<HttpRequestMessage, HttpResponseMessage> _responder;
+            public StubHandler(Func<HttpRequestMessage, HttpResponseMessage> responder) => _responder = responder;
+
+            protected override Task<HttpResponseMessage> SendAsync(HttpRequestMessage request, CancellationToken cancellationToken)
+                => Task.FromResult(_responder(request));
+        }
+
+        private static BackendApiClient BuildClient(HttpStatusCode status, string body, string mediaType)
+        {
+            var handler = new StubHandler(_ => new HttpResponseMessage(status)
+            {
+                Content = new StringContent(body, Encoding.UTF8, mediaType)
+            });
+            return new BackendApiClient(
+                new HttpClient(handler),
+                baseUrl: "https://backend.example",
+                manufacturer: "Contoso",
+                model: "TestBox",
+                serialNumber: "SN123",
+                useBootstrapTokenAuth: false,
+                bootstrapToken: null,
+                agentVersion: "2.0.0",
+                logger: null);
+        }
+
+        [Fact]
+        public async Task Backend_json_403_is_reported_as_device_not_authorized()
+        {
+            using var client = BuildClient(
+                HttpStatusCode.Forbidden,
+                "{\"success\":false,\"message\":\"Device validation failed\"}",
+                "application/json");
+
+            var ex = await Assert.ThrowsAsync<BackendAuthException>(() => client.GetAgentConfigAsync("tenant-1"));
+
+            Assert.Equal(403, ex.StatusCode);
+            Assert.False(ex.EndpointUnavailable);
+            Assert.Contains("The device is not authorized", ex.Message);
+        }
+
+        [Fact]
+        public async Task Platform_html_403_is_reported_as_endpoint_unavailable_with_page_title()
+        {
+            using var client = BuildClient(
+                HttpStatusCode.Forbidden,
+                "<!DOCTYPE html><html><head><title>Web App - Unavailable</title></head><body>stopped</body></html>",
+                "text/html");
+
+            var ex = await Assert.ThrowsAsync<BackendAuthException>(() => client.GetAgentConfigAsync("tenant-1"));
+
+            Assert.Equal(403, ex.StatusCode);
+            Assert.True(ex.EndpointUnavailable);
+            Assert.Contains("Web App - Unavailable", ex.Message);
+            Assert.Contains("NOT a device-authorization failure", ex.Message);
+            Assert.DoesNotContain("Check client certificate", ex.Message);
+        }
+
+        [Fact]
+        public async Task Html_body_with_generic_content_type_is_still_sniffed_as_platform_page()
+        {
+            using var client = BuildClient(
+                HttpStatusCode.Forbidden,
+                "  <html><body>Error 403 - This web app is stopped.</body></html>",
+                "text/plain");
+
+            var ex = await Assert.ThrowsAsync<BackendAuthException>(() => client.GetAgentConfigAsync("tenant-1"));
+
+            Assert.True(ex.EndpointUnavailable);
+            Assert.Contains("HTML platform error page", ex.Message);
+        }
+
+        [Fact]
+        public async Task Backend_401_keeps_authorization_message_and_status()
+        {
+            using var client = BuildClient(
+                HttpStatusCode.Unauthorized,
+                "{\"success\":false,\"message\":\"certificate rejected\"}",
+                "application/json");
+
+            var ex = await Assert.ThrowsAsync<BackendAuthException>(() => client.GetAgentConfigAsync("tenant-1"));
+
+            Assert.Equal(401, ex.StatusCode);
+            Assert.False(ex.EndpointUnavailable);
+            Assert.Contains("The device is not authorized", ex.Message);
+        }
+    }
+}

src/Agent/AutopilotMonitor.Agent.V2.Core/Configuration/AgentConfiguration.cs

@@ -109,6 +109,14 @@ namespace AutopilotMonitor.Agent.V2.Core.Configuration
         /// </summary>
         public bool EnableTimezoneAutoSet { get; set; } = false;
 
+        /// <summary>
+        /// Continue-Anyway observation mode (operator-set tenant opt-in, delivered via
+        /// remote config): stamped into the esp_config_detected decision-signal payload so
+        /// the DecisionEngine defangs a Device-phase ESP terminal failure into a bounded
+        /// 60-min observation instead of failing immediately. Default: false.
+        /// </summary>
+        public bool EnableEspContinueAnywayObservation { get; set; } = false;
+
         /// <summary>
         /// Optional custom path to IME logs directory for testing.
         /// If set, overrides the default %ProgramData%\Microsoft\IntuneManagementExtension\Logs path.
@@ -272,9 +280,10 @@ namespace AutopilotMonitor.Agent.V2.Core.Configuration
         /// CloudDomainJoin keys and re-probes on every change until either resolution
         /// or this timeout. 0 = no wait, fast-fail on miss (legacy behaviour).
         /// <para>
-        /// Set via the install-time CLI flag <c>--tenant-id-wait &lt;sec&gt;</c>, persisted
-        /// to <c>bootstrap-config.json</c>. The PS1 bootstrap script owns the production
-        /// default; the agent itself never assumes one.
+        /// Set via the CLI flag <c>--tenant-id-wait &lt;sec&gt;</c> or a persisted
+        /// <c>bootstrap-config.json</c> value; when neither supplies a value the agent
+        /// applies its own default of 600 s in BuildAgentConfiguration. The 0 default
+        /// here is only the POCO baseline, not the effective runtime default.
         /// </para>
         /// </summary>
         public int TenantIdWaitSeconds { get; set; } = 0;

src/Agent/AutopilotMonitor.Agent.V2.Core/Configuration/BootstrapConfigFile.cs

@@ -16,8 +16,11 @@ namespace AutopilotMonitor.Agent.V2.Core.Configuration
         /// <summary>
         /// TenantId-wait timeout in seconds, persisted from the <c>--tenant-id-wait</c>
         /// install-time arg. Read by AgentBootstrap when the registry probe finds no
-        /// TenantId on first try. 0 / missing = no wait, legacy fast-fail.
+        /// TenantId on first try. 0 = explicit opt-out (legacy fast-fail).
+        /// Null / missing (files written before this property existed) = not configured —
+        /// the agent-side default (600 s) applies. Nullable so a legacy file without the
+        /// property cannot masquerade as an explicit 0 and silently disable the wait.
         /// </summary>
-        public int TenantIdWaitSeconds { get; set; }
+        public int? TenantIdWaitSeconds { get; set; }
     }
 }

src/Agent/AutopilotMonitor.Agent.V2.Core/Configuration/RemoteConfigMerger.cs

@@ -103,6 +103,10 @@ namespace AutopilotMonitor.Agent.V2.Core.Configuration
 
             agentConfig.SendTraceEvents = remoteConfig.SendTraceEvents;
 
+            // ---------------- Continue-Anyway observation — remote wins (operator-set tenant
+            //                  opt-in; false must clear a stale cached true).
+            agentConfig.EnableEspContinueAnywayObservation = remoteConfig.EnableEspContinueAnywayObservation;
+
             // ---------------- UnrestrictedMode audit — track old/new for the caller so the
             //                  agent_unrestricted_mode_changed event can fire AFTER the
             //                  orchestrator emitter is alive (V1 parity with

src/Agent/AutopilotMonitor.Agent.V2.Core/Configuration/RemoteConfigService.cs

@@ -155,7 +155,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Configuration
                     // see the AuthCertificateRejected/DeviceNotRegistered distress signal
                     // without 100 s of needless retry delay.
                     _logger.Warning($"Config fetch authentication failed: {ex.Message}");
-                    _authFailureTracker?.RecordFailure(ex.StatusCode, "agent/config");
+                    _authFailureTracker?.RecordFailure(ex.StatusCode, "agent/config", ex.EndpointUnavailable);
                     lastException = ex;
                     LastFetchAuthStatusCode = ex.StatusCode;
                     break;
@@ -328,6 +328,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Configuration
                 EnableGeoLocation = true,
                 EnableImeMatchLog = false,
                 EnableGatherRuleDebugLog = false,
+                EnableEspContinueAnywayObservation = false,
                 Collectors = CollectorConfiguration.CreateDefault(),
                 GatherRules = new System.Collections.Generic.List<GatherRule>(),
                 ImeLogPatterns = new System.Collections.Generic.List<ImeLogPattern>()

src/Agent/AutopilotMonitor.Agent.V2.Core/Logging/AgentLogger.cs

@@ -197,7 +197,11 @@ namespace AutopilotMonitor.Agent.V2.Core.Logging
         /// <summary>
         /// Sanitizes log messages to prevent log injection attacks.
         /// Escapes control characters that could forge log entries.
-        /// Redaction patterns can be added here later as needed.
+        /// Transliterates typographic punctuation to ASCII: the log file is written as
+        /// UTF-8 without BOM, and customer-side viewers (e.g. Outlook web attachment
+        /// preview) guess cp1252 and render an em dash as mojibake ("—"). Only our own
+        /// punctuation is mapped — genuine non-ASCII payload (device names, paths) is
+        /// kept verbatim. Redaction patterns can be added here later as needed.
         /// </summary>
         private static string SanitizeLogMessage(string message)
         {
@@ -206,6 +210,17 @@ namespace AutopilotMonitor.Agent.V2.Core.Logging
             // Escape newlines and carriage returns to prevent log injection
             message = message.Replace("\r", "\\r").Replace("\n", "\\n");
 
+            message = message
+                .Replace('\u2014', '-')      // — em dash
+                .Replace('\u2013', '-')      // – en dash
+                .Replace("\u2192", "->")     // → arrow
+                .Replace("\u2026", "...")    // … ellipsis
+                .Replace('\u2018', '\'')     // ' left single quote
+                .Replace('\u2019', '\'')     // ' right single quote
+                .Replace('\u201C', '"')      // " left double quote
+                .Replace('\u201D', '"')      // " right double quote
+                .Replace('\u00A0', ' ');     // non-breaking space
+
             return message;
         }
     }

src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Runtime/SelfUpdater.cs

@@ -88,6 +88,18 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
         /// </summary>
         public static string BackendExpectedSha256 { get; set; }
 
+        /// <summary>
+        /// True once a <see cref="CheckAndApplyUpdateAsync"/> run actually obtained a version
+        /// from the manifest (regardless of whether an update was then needed/applied).
+        /// <para>
+        /// Consumed by the <c>--await-enrollment</c> path: the startup check runs pre-enrollment
+        /// where OOBE frequently has no network yet, so a failed check there says nothing about
+        /// the latest version. Once the MDM certificate appears the device provably has
+        /// connectivity — the bootstrap retries the check exactly when this is still false.
+        /// </para>
+        /// </summary>
+        public static bool LastVersionCheckSucceeded { get; private set; }
+
         /// <summary>
         /// Optional: graceful-shutdown request hook, set by the runtime host once its
         /// shutdown machinery is armed (before the orchestrator starts) and cleared in its
@@ -130,14 +142,21 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
         /// to cover hotfixes that reuse the same version number.
         /// </param>
         /// <param name="triggerReason">
-        /// "startup" (default) or "runtime_hash_mismatch". Recorded in the marker file so the
-        /// agent_version_check event (outcome=updated) can distinguish the two paths.
+        /// "startup" (default), "await_enrollment" (post-cert retry in the bootstrap), or
+        /// "runtime_hash_mismatch". Recorded in the marker file so the agent_version_check
+        /// event (outcome=updated) can distinguish the paths.
         /// </param>
         /// <param name="downloadTimeoutMsOverride">
         /// Overrides the ZIP download timeout. Startup path leaves null (fast-fail, 10s) to preserve
         /// Autopilot boot speed. Runtime trigger passes 60000 because monitoring is already running
         /// and we know an update is required.
         /// </param>
+        /// <param name="versionCheckTimeoutMsOverride">
+        /// Overrides the version-v2.json fetch timeout (default 2500ms, tuned for boot speed).
+        /// The await-enrollment retry passes a relaxed value: enrollment just completed, the
+        /// device provably has network, and a stranded old build's last self-rescue chance is
+        /// worth a few extra seconds before session registration.
+        /// </param>
         /// <param name="allowDowngrade">
         /// When false (default), a latest version strictly lower than <paramref name="currentVersion"/>
         /// is rejected even when <paramref name="forceUpdate"/> is true — prevents the production
@@ -151,7 +170,8 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
             bool forceUpdate = false,
             string triggerReason = "startup",
             int? downloadTimeoutMsOverride = null,
-            bool allowDowngrade = false)
+            bool allowDowngrade = false,
+            int? versionCheckTimeoutMsOverride = null)
         {
             Action<string> log = msg =>
             {
@@ -160,7 +180,11 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
             };
 
             var isStartupTrigger = string.Equals(triggerReason, "startup", StringComparison.Ordinal);
+            // The await-enrollment retry replaces the startup verdict — its markers must win so
+            // the agent_version_check event reflects the LATEST check, not the pre-network one.
+            var writeMarkers = isStartupTrigger || string.Equals(triggerReason, "await_enrollment", StringComparison.Ordinal);
             var downloadTimeoutMs = downloadTimeoutMsOverride ?? DownloadTimeoutMs;
+            var versionCheckTimeoutMs = versionCheckTimeoutMsOverride ?? VersionCheckTimeoutMs;
             var totalSw = Stopwatch.StartNew();
             long versionCheckMs = 0, downloadMs = 0, verifyMs = 0, extractMs = 0, swapMs = 0;
             long zipSizeBytes = 0;
@@ -174,16 +198,17 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
 
             try
             {
-                // Step 1: Fetch version-v2.json (2.5s timeout)
+                // Step 1: Fetch version-v2.json (default 2.5s timeout, relaxed on retry triggers)
                 var versionSw = Stopwatch.StartNew();
                 string manifestSha256;
-                (latestVersion, manifestSha256) = await GetLatestVersionAsync(log);
+                (latestVersion, manifestSha256) = await GetLatestVersionAsync(log, versionCheckTimeoutMs);
                 versionSw.Stop();
                 versionCheckMs = versionSw.ElapsedMilliseconds;
+                LastVersionCheckSucceeded = latestVersion != null;
 
                 if (latestVersion == null)
                 {
-                    if (isStartupTrigger)
+                    if (writeMarkers)
                         WriteSkipMarker("version_check_failed", currentVersion, null, "version-v2.json fetch failed or timed out", log);
                     return; // Could not determine latest version — continue with current
                 }
@@ -213,7 +238,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                     if (!forceUpdate)
                     {
                         log($"Self-update: current version {currentVersion} is up to date (latest: {latestVersion})");
-                        if (isStartupTrigger)
+                        if (writeMarkers)
                             WriteCheckedMarker(currentVersion, latestVersion, versionCheckMs, log);
                         return;
                     }
@@ -230,7 +255,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                     if (!forceUpdate)
                     {
                         log($"Self-update: cannot compare versions (current={currentVersion}, latest={latestVersion}) — skipping");
-                        if (isStartupTrigger)
+                        if (writeMarkers)
                             WriteSkipMarker("version_compare_failed", currentVersion, latestVersion, "version parse failed", log);
                         return;
                     }
@@ -249,7 +274,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                 downloadMs = downloadTimerSw.ElapsedMilliseconds;
                 if (!downloadOk)
                 {
-                    if (isStartupTrigger)
+                    if (writeMarkers)
                         WriteSkipMarker("download_failed", currentVersion, latestVersion, $"ZIP download failed within {downloadTimeoutMs}ms", log);
                     return;
                 }
@@ -282,7 +307,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                 if (!integrityOk)
                 {
                     CleanupStaging(null, zipPath);
-                    if (isStartupTrigger)
+                    if (writeMarkers)
                         WriteSkipMarker("integrity_mismatch", currentVersion, latestVersion, "SHA-256 mismatch on downloaded ZIP", log);
                     return;
                 }
@@ -297,7 +322,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                 extractMs = extractSw.ElapsedMilliseconds;
                 if (!extractOk)
                 {
-                    if (isStartupTrigger)
+                    if (writeMarkers)
                         WriteSkipMarker("extract_failed", currentVersion, latestVersion, "ZIP extraction failed", log);
                     return;
                 }
@@ -310,7 +335,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                 {
                     log("Self-update: staging validation failed — AutopilotMonitor.Agent.exe not found in ZIP");
                     CleanupStaging(stagingDir, zipPath);
-                    if (isStartupTrigger)
+                    if (writeMarkers)
                         WriteSkipMarker("extract_failed", currentVersion, latestVersion, "AutopilotMonitor.Agent.exe not found in ZIP", log);
                     return;
                 }
@@ -323,7 +348,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                 if (!swapOk)
                 {
                     CleanupStaging(stagingDir, zipPath);
-                    if (isStartupTrigger)
+                    if (writeMarkers)
                         WriteSkipMarker("swap_failed", currentVersion, latestVersion, "file swap failed", log);
                     return;
                 }
@@ -355,7 +380,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
             catch (Exception ex)
             {
                 log($"Self-update: unexpected error — {ex.Message}. Continuing with current version.");
-                if (isStartupTrigger)
+                if (writeMarkers)
                 {
                     try { WriteSkipMarker("unexpected_error", currentVersion, latestVersion, ex.Message, log); } catch { }
                 }
@@ -366,7 +391,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
         /// Fetches version-v2.json from blob storage and returns the version string and optional SHA-256 hash.
         /// Returns (null, null) if the check fails or times out.
         /// </summary>
-        private static async Task<(string version, string sha256)> GetLatestVersionAsync(Action<string> log)
+        private static async Task<(string version, string sha256)> GetLatestVersionAsync(Action<string> log, int timeoutMs)
         {
             try
             {
@@ -375,7 +400,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
                 var versionUrl = $"{Constants.AgentDownloadBaseUrl}/{Constants.AgentVersionFileNameForLine(2)}";
 
                 using (var handler = new HttpClientHandler())
-                using (var client = new HttpClient(handler) { Timeout = TimeSpan.FromMilliseconds(VersionCheckTimeoutMs) })
+                using (var client = new HttpClient(handler) { Timeout = TimeSpan.FromMilliseconds(timeoutMs) })
                 {
                     var json = await client.GetStringAsync(versionUrl);
                     var obj = JObject.Parse(json);
@@ -396,7 +421,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Runtime
             }
             catch (TaskCanceledException)
             {
-                log($"Self-update: version check timed out ({VersionCheckTimeoutMs}ms) — skipping update");
+                log($"Self-update: version check timed out ({timeoutMs}ms) — skipping update");
                 return (null, null);
             }
             catch (HttpRequestException ex) when (ex.Message.Contains("404"))

src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Telemetry/DeviceInfo/DeviceInfoCollector.NetworkAndSecurity.cs

@@ -1021,6 +1021,10 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.DeviceInfo
                     .ToString(System.Globalization.CultureInfo.InvariantCulture);
             if (snapshot.AllowContinueAnyway.HasValue)
                 payload[SignalPayloadKeys.EspAllowContinueAnyway] = snapshot.AllowContinueAnyway.Value ? "true" : "false";
+            // Tenant opt-in (remote config, not a registry fact) — only stamped when
+            // enabled; absent = default hard-fail semantics in the engine.
+            if (_espContinueAnywayObservationEnabled?.Invoke() == true)
+                payload[SignalPayloadKeys.EspContinueAnywayObservationEnabled] = "true";
 
             var derivationInputs = new Dictionary<string, string>(payload, StringComparer.Ordinal)
             {

src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Telemetry/DeviceInfo/DeviceInfoCollector.cs

@@ -35,6 +35,11 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.DeviceInfo
         private readonly IClock _clock;
         private readonly Persistence.StartupEventGate _startupGate;
 
+        // Live accessor for AgentConfiguration.EnableEspContinueAnywayObservation — read at
+        // esp_config_detected stamping time so a late remote-config merge takes effect on
+        // the phase-triggered re-emissions. Null (tests / legacy callers) means disabled.
+        private readonly Func<bool> _espContinueAnywayObservationEnabled;
+
         private const string RegKeyWindowsCurrentVersion = @"HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion";
 
         // Restart dedup (StartupEventGate) — exemptions. boot_time is genuinely new per boot
@@ -65,7 +70,8 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.DeviceInfo
             AgentLogger logger,
             ISignalIngressSink signalIngress = null,
             IClock clock = null,
-            Persistence.StartupEventGate startupGate = null)
+            Persistence.StartupEventGate startupGate = null,
+            Func<bool> espContinueAnywayObservationEnabled = null)
         {
             _sessionId = sessionId ?? throw new ArgumentNullException(nameof(sessionId));
             _tenantId  = tenantId  ?? throw new ArgumentNullException(nameof(tenantId));
@@ -74,6 +80,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.DeviceInfo
             _signalIngress = signalIngress;
             _clock = clock;
             _startupGate = startupGate;
+            _espContinueAnywayObservationEnabled = espContinueAnywayObservationEnabled;
         }
 
         /// <summary>

src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Telemetry/Gather/GatherRuleGuards.cs

@@ -3,6 +3,7 @@ using System.Collections.Generic;
 using System.IO;
 using System.Linq;
 using System.Reflection;
+using System.Text.RegularExpressions;
 using Newtonsoft.Json.Linq;
 
 namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.Gather
@@ -35,6 +36,24 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.Gather
         public static readonly IReadOnlyCollection<string> AllowedCommands;
         public static readonly IReadOnlyList<string> AllowedEventLogChannels;
 
+        // Derived from AllowedWmiQueryPrefixes: entries of the canonical form
+        // "SELECT * FROM <class>" contribute their class name; any other entry
+        // (none today) stays on the legacy literal-prefix path.
+        private static readonly HashSet<string> AllowedWmiClasses;
+        private static readonly List<string> ResidualWmiPrefixes;
+
+        // Canonical allowlist entry: exactly "SELECT * FROM <class>".
+        private static readonly Regex CanonicalWmiEntryRegex = new Regex(
+            @"^SELECT\s+\*\s+FROM\s+([A-Za-z_][A-Za-z0-9_]*)$",
+            RegexOptions.IgnoreCase | RegexOptions.Compiled);
+
+        // WQL shape "SELECT <* | property-list> FROM <class>", class token bounded
+        // by whitespace or end of string (a trailing WHERE clause is permitted,
+        // "Win32_BIOSX" cannot pass as "Win32_BIOS").
+        private static readonly Regex WmiQueryRegex = new Regex(
+            @"^SELECT\s+(?:\*|[A-Za-z_][A-Za-z0-9_]*(?:\s*,\s*[A-Za-z_][A-Za-z0-9_]*)*)\s+FROM\s+([A-Za-z_][A-Za-z0-9_]*)(?=\s|$)",
+            RegexOptions.IgnoreCase | RegexOptions.Compiled);
+
         // Hard blocks: never allowed, even in unrestricted mode
         private static readonly string BlockedUsersPrefix = Path.GetFullPath(@"C:\Users");
         private static readonly string[] AdditionalHardBlockedPathPrefixes = new[]
@@ -101,6 +120,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.Gather
                     AllowedCommands = new HashSet<string>(commands, StringComparer.OrdinalIgnoreCase);
 
                     AllowedEventLogChannels = FlattenCategorized(obj, "eventLogChannels", "channels");
+                    DeriveWmiClassAllowlist(AllowedWmiQueryPrefixes, out AllowedWmiClasses, out ResidualWmiPrefixes);
                     return;
                 }
             }
@@ -115,6 +135,26 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.Gather
             AllowedWmiQueryPrefixes = new List<string>();
             AllowedCommands = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
             AllowedEventLogChannels = new List<string>();
+            AllowedWmiClasses = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
+            ResidualWmiPrefixes = new List<string>();
+        }
+
+        private static void DeriveWmiClassAllowlist(
+            IReadOnlyList<string> entries,
+            out HashSet<string> classes,
+            out List<string> residual)
+        {
+            classes = new HashSet<string>(StringComparer.OrdinalIgnoreCase);
+            residual = new List<string>();
+
+            foreach (var entry in entries)
+            {
+                var match = CanonicalWmiEntryRegex.Match(entry.Trim());
+                if (match.Success)
+                    classes.Add(match.Groups[1].Value);
+                else
+                    residual.Add(entry);
+            }
         }
 
         private static string LoadEmbeddedGuardrails()
@@ -250,8 +290,9 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.Gather
         }
 
         /// <summary>
-        /// Returns true if the WMI query (trimmed) matches an allowed prefix
-        /// with boundary matching (next char must be whitespace or end of string).
+        /// Returns true if the WMI query (trimmed) is "SELECT &lt;* | property-list&gt; FROM &lt;class&gt;"
+        /// (optionally followed by a WHERE clause) against a class on the allowlist.
+        /// A property projection exposes a strict subset of the allowed "SELECT *".
         /// </summary>
         public static bool IsWmiQueryAllowed(string query)
             => IsWmiQueryAllowed(query, unrestrictedMode: false);
@@ -269,7 +310,13 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Telemetry.Gather
                 return true;
 
             var trimmed = query.Trim();
-            return AllowedWmiQueryPrefixes.Any(prefix =>
+
+            var match = WmiQueryRegex.Match(trimmed);
+            if (match.Success && AllowedWmiClasses.Contains(match.Groups[1].Value))
+                return true;
+
+            // Non-canonical allowlist entries keep the legacy literal-prefix match.
+            return ResidualWmiPrefixes.Any(prefix =>
                 trimmed.StartsWith(prefix, StringComparison.OrdinalIgnoreCase) &&
                 (trimmed.Length == prefix.Length || char.IsWhiteSpace(trimmed[prefix.Length])));
         }

src/Agent/AutopilotMonitor.Agent.V2.Core/Monitoring/Transport/BackendApiClient.cs

@@ -137,7 +137,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Transport
                             continue;
                         }
 
-                        ThrowOnAuthFailure(response);
+                        await ThrowOnAuthFailureAsync(response).ConfigureAwait(false);
                         response.EnsureSuccessStatusCode();
 
                         var responseJson = await response.Content.ReadAsStringAsync().ConfigureAwait(false);
@@ -236,7 +236,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Transport
 
                 using (response)
                 {
-                    ThrowOnAuthFailure(response);
+                    await ThrowOnAuthFailureAsync(response).ConfigureAwait(false);
                     response.EnsureSuccessStatusCode();
 
                     var responseJson = await response.Content.ReadAsStringAsync().ConfigureAwait(false);
@@ -282,16 +282,97 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Transport
         /// <summary>
         /// Throws <see cref="BackendAuthException"/> for 401/403 responses so callers can
         /// distinguish authentication failures from transient server errors.
+        /// <para>
+        /// Field case (sits-d.cloud, 2026-08-09): a pre-cutover agent build pointed at a
+        /// retired Function App; Azure answers for a stopped app with a platform-level 403
+        /// carrying an HTML error page ("Web App - Unavailable"). Our backend never returns
+        /// HTML on 401/403 — an HTML body therefore means the response did not come from the
+        /// backend's auth layer at all (stopped/retired app, edge proxy, WAF). Reporting that
+        /// as "device is not authorized" sends troubleshooting down the wrong path, so the
+        /// two cases get distinct messages and the exception carries a flag.
+        /// </para>
         /// </summary>
-        private static void ThrowOnAuthFailure(HttpResponseMessage response)
+        private static async Task ThrowOnAuthFailureAsync(HttpResponseMessage response)
         {
-            if (response.StatusCode == System.Net.HttpStatusCode.Unauthorized ||
-                response.StatusCode == System.Net.HttpStatusCode.Forbidden)
+            if (response.StatusCode != System.Net.HttpStatusCode.Unauthorized &&
+                response.StatusCode != System.Net.HttpStatusCode.Forbidden)
             {
+                return;
+            }
+
+            var statusText = $"{(int)response.StatusCode} {response.StatusCode}";
+
+            string body = null;
+            try
+            {
+                body = await response.Content.ReadAsStringAsync().ConfigureAwait(false);
+            }
+            catch
+            {
+                // Body is diagnostic-only; classification below treats null as non-HTML.
+            }
+
+            if (LooksLikeHtmlErrorPage(response, body))
+            {
+                var title = TryExtractHtmlTitle(body);
+                var titleText = title != null ? $" (\"{title}\")" : string.Empty;
+                var host = response.RequestMessage?.RequestUri?.Host ?? "(unknown host)";
                 throw new BackendAuthException(
-                    $"Backend returned {(int)response.StatusCode} {response.StatusCode}. " +
-                    "The device is not authorized. Check client certificate and Autopilot device validation.",
-                    (int)response.StatusCode);
+                    $"Backend returned {statusText} with an HTML platform error page{titleText} from '{host}'. " +
+                    "This response did not come from the Autopilot Monitor backend — the endpoint appears stopped, " +
+                    "retired, or intercepted by a proxy. This is NOT a device-authorization failure; an outdated " +
+                    "agent build pointing at a stale ApiBaseUrl is the most likely cause.",
+                    (int)response.StatusCode,
+                    endpointUnavailable: true);
+            }
+
+            throw new BackendAuthException(
+                $"Backend returned {statusText}. " +
+                "The device is not authorized. Check client certificate and Autopilot device validation.",
+                (int)response.StatusCode);
+        }
+
+        /// <summary>
+        /// True when a 401/403 response is an HTML page rather than a backend JSON/plain-text
+        /// reply. Checks the declared Content-Type first, then falls back to sniffing the body
+        /// prefix so a platform page served with a generic Content-Type is still recognized.
+        /// </summary>
+        private static bool LooksLikeHtmlErrorPage(HttpResponseMessage response, string body)
+        {
+            var mediaType = response.Content?.Headers?.ContentType?.MediaType;
+            if (string.Equals(mediaType, "text/html", StringComparison.OrdinalIgnoreCase))
+                return true;
+
+            if (string.IsNullOrWhiteSpace(body))
+                return false;
+
+            var head = body.TrimStart();
+            return head.StartsWith("<!DOCTYPE", StringComparison.OrdinalIgnoreCase) ||
+                   head.StartsWith("<html", StringComparison.OrdinalIgnoreCase);
+        }
+
+        /// <summary>
+        /// Best-effort extraction of the HTML &lt;title&gt; (e.g. "Web App - Unavailable") so the
+        /// log line names the platform page without dumping markup. Null when absent/oversized.
+        /// </summary>
+        private static string TryExtractHtmlTitle(string body)
+        {
+            if (string.IsNullOrEmpty(body)) return null;
+            try
+            {
+                var match = System.Text.RegularExpressions.Regex.Match(
+                    body, @"<title[^>]*>\s*(.*?)\s*</title>",
+                    System.Text.RegularExpressions.RegexOptions.IgnoreCase |
+                    System.Text.RegularExpressions.RegexOptions.Singleline,
+                    TimeSpan.FromMilliseconds(250));
+                if (!match.Success) return null;
+
+                var title = match.Groups[1].Value;
+                return title.Length == 0 ? null : (title.Length > 80 ? title.Substring(0, 80) : title);
+            }
+            catch
+            {
+                return null;
             }
         }
 
@@ -330,7 +411,9 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Transport
     }
 
     /// <summary>
-    /// Thrown when the backend returns 401 or 403, indicating the device is not authorized.
+    /// Thrown when the backend returns 401 or 403, indicating the device is not authorized —
+    /// or, when <see cref="EndpointUnavailable"/> is set, that the 401/403 was a platform-level
+    /// HTML response (stopped/retired app, edge proxy) rather than a backend auth verdict.
     /// </summary>
     public class BackendAuthException : Exception
     {
@@ -340,9 +423,16 @@ namespace AutopilotMonitor.Agent.V2.Core.Monitoring.Transport
         /// </summary>
         public int StatusCode { get; }
 
-        public BackendAuthException(string message, int statusCode = 0) : base(message)
+        /// <summary>
+        /// True when the 401/403 carried an HTML platform error page — the endpoint itself is
+        /// stopped/retired/proxied and the status says nothing about this device's authorization.
+        /// </summary>
+        public bool EndpointUnavailable { get; }
+
+        public BackendAuthException(string message, int statusCode = 0, bool endpointUnavailable = false) : base(message)
         {
             StatusCode = statusCode;
+            EndpointUnavailable = endpointUnavailable;
         }
     }
 }

src/Agent/AutopilotMonitor.Agent.V2.Core/Orchestration/DefaultComponentFactory.cs

@@ -127,7 +127,10 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
                 ingress: ingress,
                 clock: clock,
                 logger: logger,
-                startupGate: _startupEventGate));
+                startupGate: _startupEventGate,
+                // Live accessor (not the current value): a remote-config merge landing after
+                // factory construction must still reach the phase-triggered re-emissions.
+                espContinueAnywayObservationEnabled: () => _agentConfig.EnableEspContinueAnywayObservation));
 
             // Session caa6cf50 gate-starvation fix: the EspAndHelloTracker's user-ESP-apps-settled
             // probe reads the IME log host, which is constructed further down. Same closure-over-

src/Agent/AutopilotMonitor.Agent.V2.Core/Orchestration/EnrollmentOrchestrator.cs

@@ -91,6 +91,11 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
         // (tests / legacy callers) means trace events always flow.
         private readonly Func<bool>? _traceEventsEnabled;
 
+        // Live accessor for AgentConfiguration.EnableEspContinueAnywayObservation — read at
+        // esp_config_detected stamping time so a remote-config merge that lands after
+        // construction still takes effect. Null (tests / legacy callers) means disabled.
+        private readonly Func<bool>? _espContinueAnywayObservationEnabled;
+
         // Max-lifetime watchdog (M4.6.α). Timer is armed in Start() when _agentMaxLifetime != null.
         private System.Threading.Timer? _maxLifetimeTimer;
         private int _terminatedFired;
@@ -173,7 +178,8 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
             TimeSpan? agentMaxLifetime = null,
             int uploadBatchSize = 100,
             IDeadlineScheduler? schedulerOverride = null,
-            Func<bool>? traceEventsEnabled = null)
+            Func<bool>? traceEventsEnabled = null,
+            Func<bool>? espContinueAnywayObservationEnabled = null)
         {
             if (string.IsNullOrEmpty(sessionId)) throw new ArgumentException("SessionId is mandatory.", nameof(sessionId));
             if (string.IsNullOrEmpty(tenantId)) throw new ArgumentException("TenantId is mandatory.", nameof(tenantId));
@@ -214,6 +220,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
 
             _schedulerOverride = schedulerOverride;
             _traceEventsEnabled = traceEventsEnabled;
+            _espContinueAnywayObservationEnabled = espContinueAnywayObservationEnabled;
         }
 
         /// <summary>
@@ -773,6 +780,10 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
                         .ToString(System.Globalization.CultureInfo.InvariantCulture);
                 if (snapshot.AllowContinueAnyway.HasValue)
                     payload[SignalPayloadKeys.EspAllowContinueAnyway] = snapshot.AllowContinueAnyway.Value ? "true" : "false";
+                // Tenant opt-in (remote config, not a registry fact) — only stamped when
+                // enabled; absent = default hard-fail semantics in the engine.
+                if (_espContinueAnywayObservationEnabled?.Invoke() == true)
+                    payload[SignalPayloadKeys.EspContinueAnywayObservationEnabled] = "true";
 
                 _ingress!.Post(
                     kind: DecisionSignalKind.EspConfigDetected,

src/Agent/AutopilotMonitor.Agent.V2.Core/Orchestration/Hosts/DeviceInfoHost.cs

@@ -82,7 +82,8 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
             ISignalIngressSink ingress,
             IClock clock,
             AgentLogger logger,
-            Persistence.StartupEventGate? startupGate = null)
+            Persistence.StartupEventGate? startupGate = null,
+            Func<bool>? espContinueAnywayObservationEnabled = null)
         {
             if (ingress == null) throw new ArgumentNullException(nameof(ingress));
             if (clock == null) throw new ArgumentNullException(nameof(clock));
@@ -92,7 +93,8 @@ namespace AutopilotMonitor.Agent.V2.Core.Orchestration
             // when it reads the FirstSync SkipUser/SkipDevice registry values, so that Fix 8's
             // reducer guards have the SkipUserEsp/SkipDeviceEsp state facts to read.
             _collector = new Monitoring.Telemetry.DeviceInfo.DeviceInfoCollector(
-                sessionId, tenantId, post, logger, ingress, clock, startupGate);
+                sessionId, tenantId, post, logger, ingress, clock, startupGate,
+                espContinueAnywayObservationEnabled);
             _observableIngress = ingress as SignalIngress;
         }
 

src/Agent/AutopilotMonitor.Agent.V2.Core/Security/AuthFailureTracker.cs

@@ -46,6 +46,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Security
         private int _consecutiveFailures;
         private DateTime? _firstFailureUtc;
         private int _thresholdFired;           // 0/1 — Interlocked exchange makes the event single-shot
+        private int _distressDispatched;       // 0/1 per failure streak — reset by RecordSuccess
 
         public AuthFailureTracker(
             int maxFailures,
@@ -77,18 +78,27 @@ namespace AutopilotMonitor.Agent.V2.Core.Security
         public void RecordSuccess()
         {
             if (Interlocked.Exchange(ref _consecutiveFailures, 0) == 0) return; // no-op if already zero
+            Interlocked.Exchange(ref _distressDispatched, 0);
             lock (_windowLock) { _firstFailureUtc = null; }
         }
 
         /// <summary>
-        /// Record a 401/403 from the given <paramref name="operation"/>. V1 parity — on the FIRST
-        /// failure this fires a single distress report via the constructor-injected
-        /// <see cref="DistressReporter"/> (401 → AuthCertificateRejected, 403 → DeviceNotRegistered).
-        /// Subsequent failures are logged + counted but do not dispatch further distress signals.
-        /// Emits <see cref="ThresholdExceeded"/> the first time either ceiling is crossed;
-        /// subsequent calls after termination is armed are no-ops.
+        /// Record a 401/403 from the given <paramref name="operation"/>. V1 parity — the first
+        /// qualifying failure of a streak fires a single distress report via the
+        /// constructor-injected <see cref="DistressReporter"/> (401 → AuthCertificateRejected,
+        /// 403 → DeviceNotRegistered). Subsequent failures are logged + counted but do not
+        /// dispatch further distress signals. Emits <see cref="ThresholdExceeded"/> the first
+        /// time either ceiling is crossed; subsequent calls after termination is armed are no-ops.
         /// </summary>
-        public void RecordFailure(int statusCode, string operation)
+        /// <param name="endpointUnavailable">
+        /// True when the 401/403 was a platform-level HTML response (stopped/retired app, edge
+        /// proxy — see <see cref="BackendAuthException.EndpointUnavailable"/>). Suppresses the
+        /// distress dispatch: the report would go to the very endpoint that just answered with a
+        /// platform error page, and its DeviceNotRegistered classification would be wrong anyway.
+        /// Counting and the shutdown thresholds are unaffected. Should a later failure in the
+        /// same streak come from a live backend, that one still dispatches the streak's report.
+        /// </param>
+        public void RecordFailure(int statusCode, string operation, bool endpointUnavailable = false)
         {
             if (Volatile.Read(ref _thresholdFired) == 1) return;
 
@@ -102,10 +112,21 @@ namespace AutopilotMonitor.Agent.V2.Core.Security
                 firstFailureAt = _firstFailureUtc.Value;
             }
 
-            // V1 parity — only the first consecutive failure produces a distress report. Repeat
-            // hits at 2..N use the local log only; the backend already knows from the first
-            // report that the device is unauthorized.
-            if (count == 1 && _distressReporter != null)
+            // V1 parity — only one distress report per failure streak. Repeat hits use the local
+            // log only; the backend already knows from the first report that the device is
+            // unauthorized. Endpoint-unavailable failures never dispatch (nothing to send to) but
+            // also never consume the streak's report slot.
+            if (endpointUnavailable)
+            {
+                if (_distressReporter != null && Volatile.Read(ref _distressDispatched) == 0)
+                {
+                    _logger.Info(
+                        $"AuthFailureTracker: distress report suppressed for {operation} (http {statusCode}) — " +
+                        "endpoint unavailable (platform error page), a report cannot reach a stopped endpoint.");
+                }
+            }
+            else if (_distressReporter != null &&
+                     Interlocked.CompareExchange(ref _distressDispatched, 1, 0) == 0)
             {
                 var distressType = statusCode == 403
                     ? DistressErrorType.DeviceNotRegistered

src/Agent/AutopilotMonitor.Agent.V2.Core/Security/SessionRegistrationHelper.cs

@@ -89,7 +89,7 @@ namespace AutopilotMonitor.Agent.V2.Core.Security
                     // Feed the central auth-failure tracker so the first-failure distress dispatches
                     // and the shutdown threshold advances. V1 parity: no retry on auth-failure —
                     // backend has definitively rejected the cert/token.
-                    authFailureTracker?.RecordFailure(ex.StatusCode, "agent/register-session");
+                    authFailureTracker?.RecordFailure(ex.StatusCode, "agent/register-session", ex.EndpointUnavailable);
                     return SessionRegistrationResult.AuthFailed(ex.StatusCode, ex.Message);
                 }
                 catch (Exception ex)

src/Agent/AutopilotMonitor.Agent.V2/Program.InstallMode.cs

@@ -135,7 +135,7 @@ namespace AutopilotMonitor.Agent.V2
                         $"{(bootstrapTokenGiven ? "*" : "")}, " +
                         $"tenantId={(string.IsNullOrEmpty(merged.TenantId) ? "no" : "yes")}" +
                         $"{(tenantIdGiven ? "*" : "")}, " +
-                        $"tenantIdWait={merged.TenantIdWaitSeconds}s" +
+                        $"tenantIdWait={(merged.TenantIdWaitSeconds is int w ? $"{w}s" : "default")}" +
                         $"{(tenantIdWaitGiven ? "*" : "")})." +
                         " Star = explicitly set on this --install invocation, others were preserved from prior config.");
                 }
@@ -361,9 +361,12 @@ namespace AutopilotMonitor.Agent.V2
             {
                 BootstrapToken = bootstrapTokenGiven ? bootstrapTokenArg : existing?.BootstrapToken,
                 TenantId = tenantIdGiven ? tenantIdArg : existing?.TenantId,
+                // Not given → carry the prior value through, INCLUDING null. Persisting a
+                // number here would freeze the agent-side default into the file; persisting
+                // null keeps "not configured" distinguishable from an explicit opt-out (0).
                 TenantIdWaitSeconds = tenantIdWaitGiven
                     ? tenantIdWaitSeconds
-                    : (existing?.TenantIdWaitSeconds ?? 600),
+                    : existing?.TenantIdWaitSeconds,
             };
         }
 

src/Agent/AutopilotMonitor.Agent.V2/Program.cs

@@ -339,13 +339,15 @@ namespace AutopilotMonitor.Agent.V2
             // over the agent-side default (600 s). Hybrid-AAD-joined devices typically
             // need ~5 min for the AAD device cert to land in the registry; 600 s leaves
             // headroom. Pass `--tenant-id-wait 0` to opt out (legacy fast-fail).
+            // A bootstrap-config.json whose TenantIdWaitSeconds is null (file predates the
+            // property) is "not configured", NOT an opt-out — the default must still apply.
             var tenantIdWaitRaw = GetArgValue(args, "--tenant-id-wait");
             const int tenantIdWaitDefaultSeconds = 600;
             int tenantIdWaitSeconds = tenantIdWaitDefaultSeconds;
             if (!string.IsNullOrEmpty(tenantIdWaitRaw) && int.TryParse(tenantIdWaitRaw, out var parsedTenantIdWait))
                 tenantIdWaitSeconds = parsedTenantIdWait;
-            else if (bootstrapConfig != null)
-                tenantIdWaitSeconds = bootstrapConfig.TenantIdWaitSeconds;
+            else if (bootstrapConfig?.TenantIdWaitSeconds != null)
+                tenantIdWaitSeconds = bootstrapConfig.TenantIdWaitSeconds.Value;
 
             var useBootstrapTokenAuth = !string.IsNullOrEmpty(bootstrapToken);
 

src/Agent/AutopilotMonitor.Agent.V2/Runtime/AgentBootstrap.cs

@@ -121,6 +121,8 @@ namespace AutopilotMonitor.Agent.V2.Runtime
 
                 // Await-enrollment is one-shot — remove the persisted config so subsequent restarts proceed normally.
                 Program.DeleteAwaitEnrollmentConfig(dataDirectory, logger);
+
+                RetrySelfUpdateAfterEnrollment(logger, consoleMode);
             }
 
             // Event-driven TenantId wait — kicks in only when the user opted-in via
@@ -152,6 +154,46 @@ namespace AutopilotMonitor.Agent.V2.Runtime
                 isWhiteGloveResume: isWhiteGloveResume,
                 cleanupServiceFactory: cleanupServiceFactory);
         }
+
+        /// <summary>
+        /// One-shot self-update retry after the MDM certificate appeared. The startup check in
+        /// <c>Main</c> runs pre-enrollment, where an --await-enrollment install typically has no
+        /// network yet — its failure says nothing about the latest version. The certificate is
+        /// proof of connectivity, and this is the last chance to swap a stranded old build
+        /// (stale baked-in ApiBaseUrl) for a current one BEFORE session registration hits a
+        /// retired endpoint (field case sits-d.cloud, 2026-08-09). Timeouts are relaxed vs. the
+        /// boot-speed-tuned startup check; enrollment is already past OOBE's critical path here.
+        /// On update success the process restarts and this method never returns; the
+        /// await-enrollment config was already removed, so the new binary proceeds normally.
+        /// </summary>
+        private static void RetrySelfUpdateAfterEnrollment(AgentLogger logger, bool consoleMode)
+        {
+            if (SelfUpdater.LastVersionCheckSucceeded)
+                return; // Startup check reached the manifest — its verdict stands.
+
+            try
+            {
+                logger.Info("Await-enrollment: startup self-update check had failed (no network in early OOBE) — retrying now that enrollment proves connectivity.");
+                var agentDir = System.IO.Path.GetDirectoryName(
+                    System.Reflection.Assembly.GetExecutingAssembly().Location) ?? string.Empty;
+                SelfUpdater.CheckAndApplyUpdateAsync(
+                    currentVersion: Program.GetAgentVersion(),
+                    agentDir: agentDir,
+                    consoleMode: consoleMode,
+                    triggerReason: "await_enrollment",
+                    downloadTimeoutMsOverride: 60000,
+                    // Fresh --await-enrollment installs have no cached remote config yet, so
+                    // there is no admin downgrade override to honour on this path.
+                    allowDowngrade: false,
+                    versionCheckTimeoutMsOverride: 10000).GetAwaiter().GetResult();
+            }
+            catch (Exception ex)
+            {
+                // Same contract as the startup call site: the updater swallows its own errors;
+                // anything escaping must not break bootstrap — the current binary continues.
+                logger.Warning($"Await-enrollment: self-update retry failed unexpectedly: {ex.Message}. Continuing with current version.");
+            }
+        }
     }
 
     /// <summary>

src/Agent/AutopilotMonitor.Agent.V2/Runtime/AgentRuntimeHost.cs

@@ -135,7 +135,8 @@ namespace AutopilotMonitor.Agent.V2.Runtime
                 uploadBatchSize: uploadBatchSize,
                 // Live read: remote config is merged into agentConfig mid-session, so the
                 // trace-event gate must follow it rather than snapshot at construction.
-                traceEventsEnabled: () => agentConfig.SendTraceEvents))
+                traceEventsEnabled: () => agentConfig.SendTraceEvents,
+                espContinueAnywayObservationEnabled: () => agentConfig.EnableEspContinueAnywayObservation))
             {
                 using (var shutdown = new ManualResetEventSlim(false))
                 using (var shutdownComplete = new ManualResetEventSlim(false))

src/Agent/AutopilotMonitor.DecisionCore.Tests/ContinueAnywayObservationTests.cs

@@ -0,0 +1,383 @@
+using System;
+using System.Collections.Generic;
+using System.Linq;
+using AutopilotMonitor.DecisionCore.Engine;
+using AutopilotMonitor.DecisionCore.Signals;
+using AutopilotMonitor.DecisionCore.State;
+using Xunit;
+
+namespace AutopilotMonitor.DecisionCore.Tests
+{
+    /// <summary>
+    /// Continue-Anyway observation mode (tenant c9787ba2, session 53d1e9f6, 2026-08-08).
+    /// A Device-phase ESP terminal failure (AccountSetup never entered) on a
+    /// Continue-Anyway-enabled profile is a hard fail by default; with the tenant opt-in
+    /// (<c>EspContinueAnywayObservationEnabled</c> observation) it is defanged into a 60-min
+    /// observation advisory instead. Coverage:
+    /// <list type="bullet">
+    ///   <item>Defang + 60-min AdvisoryCompletion arming (vs the classic 30-min window).</item>
+    ///   <item>Opt-out / ContinueAnyway-off / self-deploying-scenario regressions still hard-fail.</item>
+    ///   <item>Real-user desktop + Hello-disabled completes eagerly through Finalizing; the
+    ///         terminal enrollment_complete carries the espSoftFailure marker.</item>
+    ///   <item>Window expiry without desktop un-defangs to esp_terminal_failure.</item>
+    ///   <item>Deadline fire with desktop but unknown Hello promotes to AwaitingHello.</item>
+    ///   <item>Category recovery ("Try again") resolves the advisory (classic 4910a5a5 hook).</item>
+    /// </list>
+    /// </summary>
+    public sealed class ContinueAnywayObservationTests
+    {
+        private static readonly DateTime T0 = new DateTime(2026, 8, 7, 15, 16, 0, DateTimeKind.Utc);
+
+        private static DecisionSignal MakeSignal(
+            long ordinal,
+            DecisionSignalKind kind,
+            DateTime occurredAtUtc,
+            IReadOnlyDictionary<string, string>? payload = null,
+            string sourceOrigin = "test")
+        {
+            return new DecisionSignal(
+                sessionSignalOrdinal: ordinal,
+                sessionTraceOrdinal: ordinal,
+                kind: kind,
+                kindSchemaVersion: 1,
+                occurredAtUtc: occurredAtUtc,
+                sourceOrigin: sourceOrigin,
+                evidence: new Evidence(EvidenceKind.Synthetic, $"{kind}-{ordinal}", "test"),
+                payload: payload);
+        }
+
+        private static DecisionSignal DeadlineFired(long ordinal, DateTime occurredAtUtc, string deadlineName) =>
+            MakeSignal(ordinal, DecisionSignalKind.DeadlineFired, occurredAtUtc,
+                new Dictionary<string, string> { [SignalPayloadKeys.Deadline] = deadlineName });
+
+        private static DecisionEffect SingleTimelineEffect(DecisionStep step, string eventType) =>
+            step.Effects.Single(e =>
+                e.Kind == DecisionEffectKind.EmitEventTimelineEntry
+                && e.Parameters != null
+                && e.Parameters.TryGetValue("eventType", out var et)
+                && et == eventType);
+
+        private static ActiveDeadline? FindDeadline(DecisionState state, string name) =>
+            state.Deadlines.FirstOrDefault(d => d.Name == name);
+
+        /// <summary>
+        /// Replays the session-53d1e9f6 shape up to (but excluding) the ESP terminal failure:
+        /// Continue-Anyway profile with the tenant observation opt-in stamped into the
+        /// EspConfigDetected payload, Hello policy disabled, DeviceSetup entered —
+        /// AccountSetup never reached (the 30-min Device-ESP wall kills it first).
+        /// </summary>
+        private static DecisionState SetupObservationEligibleSession(
+            DecisionEngine engine,
+            bool observationOptIn = true,
+            bool continueAnyway = true,
+            bool helloPolicyDisabled = true)
+        {
+            var state = DecisionState.CreateInitial("sess-53d1e9f6", "tenant-c9787ba2", T0);
+            state = engine.Reduce(state, MakeSignal(0, DecisionSignalKind.SessionStarted, T0)).NewState;
+            var espConfig = new Dictionary<string, string>
+            {
+                [SignalPayloadKeys.SkipUserEsp] = "false",
+                [SignalPayloadKeys.SkipDeviceEsp] = "false",
+                [SignalPayloadKeys.EspSyncFailureTimeoutMinutes] = "30",
+                [SignalPayloadKeys.EspAllowContinueAnyway] = continueAnyway ? "true" : "false",
+            };
+            if (observationOptIn)
+            {
+                espConfig[SignalPayloadKeys.EspContinueAnywayObservationEnabled] = "true";
+            }
+            state = engine.Reduce(state, MakeSignal(
+                5, DecisionSignalKind.EspConfigDetected, T0.AddMinutes(1), espConfig)).NewState;
+            if (helloPolicyDisabled)
+            {
+                state = engine.Reduce(state, MakeSignal(
+                    8, DecisionSignalKind.HelloPolicyDetected, T0.AddMinutes(1),
+                    new Dictionary<string, string> { [SignalPayloadKeys.HelloEnabled] = "false" })).NewState;
+            }
+            state = engine.Reduce(state, MakeSignal(
+                10, DecisionSignalKind.EspPhaseChanged, T0.AddMinutes(2),
+                new Dictionary<string, string> { [SignalPayloadKeys.EspPhase] = "DeviceSetup" })).NewState;
+            return state;
+        }
+
+        private static DecisionStep ApplyDeviceSetupTerminalFailure(DecisionEngine engine, DecisionState state, long ordinal = 50) =>
+            engine.Reduce(state, MakeSignal(
+                ordinal, DecisionSignalKind.EspTerminalFailure, T0.AddMinutes(38),
+                new Dictionary<string, string>
+                {
+                    ["failureType"] = "Provisioning_DeviceSetup_Certificates_Failed",
+                    ["failedSubcategory"] = "Certificates",
+                    ["category"] = "DeviceSetup",
+                }));
+
+        // ============================================================ defang path ====
+
+        [Fact]
+        public void DeviceSetupFailure_WithOptIn_DefangsToObservationAdvisory()
+        {
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            Assert.Null(state.AccountSetupEnteredUtc);
+
+            var preStage = state.Stage;
+            var step = ApplyDeviceSetupTerminalFailure(engine, state);
+
+            // No terminal — stage unchanged, advisory anchored, no LastFailureTrigger.
+            Assert.Equal(preStage, step.NewState.Stage);
+            Assert.Null(step.NewState.Outcome);
+            Assert.NotNull(step.NewState.EspAdvisoryFailureRecordedUtc);
+            Assert.Equal("DeviceSetup", step.NewState.EspAdvisoryFailureCategory!.Value);
+            Assert.Null(step.NewState.LastFailureTrigger);
+
+            var advisory = SingleTimelineEffect(step, "esp_failure_advisory");
+            Assert.Equal("esp_failure_defanged_continueanyway_observation", advisory.Parameters!["advisoryReason"]);
+            Assert.Equal("Warning", advisory.Parameters!["severity"]);
+            Assert.Equal("true", advisory.Parameters!["mayHaveContinuedAnyway"]);
+
+            Assert.DoesNotContain(step.Effects, e =>
+                e.Parameters != null
+                && e.Parameters.TryGetValue("eventType", out var et)
+                && et == "enrollment_failed");
+        }
+
+        [Fact]
+        public void DeviceSetupFailure_WithOptIn_Arms60MinObservationWindow()
+        {
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+
+            var step = ApplyDeviceSetupTerminalFailure(engine, state);
+
+            var deadline = FindDeadline(step.NewState, DeadlineNames.AdvisoryCompletion);
+            Assert.NotNull(deadline);
+            // 60-min observation window (user decision 2026-08-08) — NOT the classic 30-min
+            // advisory-completion window.
+            Assert.Equal(T0.AddMinutes(38).AddMinutes(60), deadline!.DueAtUtc);
+        }
+
+        // ==================================================== negative / regression ====
+
+        [Fact]
+        public void DeviceSetupFailure_WithoutOptIn_StillHardFails()
+        {
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine, observationOptIn: false);
+
+            var step = ApplyDeviceSetupTerminalFailure(engine, state);
+
+            Assert.Equal(SessionStage.Failed, step.NewState.Stage);
+            Assert.Equal(SessionOutcome.EnrollmentFailed, step.NewState.Outcome);
+            Assert.Null(step.NewState.EspAdvisoryFailureRecordedUtc);
+            SingleTimelineEffect(step, "enrollment_failed");
+        }
+
+        [Fact]
+        public void DeviceSetupFailure_WithOptIn_ButContinueAnywayDisabled_StillHardFails()
+        {
+            // The observation only makes sense when the ESP profile actually offers the
+            // "Continue anyway" button — without it the user cannot reach the desktop.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine, continueAnyway: false);
+
+            var step = ApplyDeviceSetupTerminalFailure(engine, state);
+
+            Assert.Equal(SessionStage.Failed, step.NewState.Stage);
+            Assert.Null(step.NewState.EspAdvisoryFailureRecordedUtc);
+            SingleTimelineEffect(step, "enrollment_failed");
+        }
+
+        [Fact]
+        public void DeviceSetupFailure_WithOptIn_OnSelfDeployingProfile_StillHardFails()
+        {
+            // Scope gate: self-deploying/kiosk flows have no interactive user who could press
+            // "Continue anyway" — observation would only ever expire. Immediate fail is truthful.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = engine.Reduce(state, MakeSignal(
+                20, DecisionSignalKind.EnrollmentFactsObserved, T0.AddMinutes(3),
+                new Dictionary<string, string> { [SignalPayloadKeys.IsSelfDeployingProfile] = "true" })).NewState;
+
+            var step = ApplyDeviceSetupTerminalFailure(engine, state);
+
+            Assert.Equal(SessionStage.Failed, step.NewState.Stage);
+            Assert.Null(step.NewState.EspAdvisoryFailureRecordedUtc);
+        }
+
+        [Fact]
+        public void ClassicAdvisory_WithAccountSetupEntered_KeepsClassicReasonAndWindow()
+        {
+            // Regression: the opt-in must not change the classic post-AccountSetup advisory —
+            // same advisoryReason literal, same 30-min window.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = engine.Reduce(state, MakeSignal(
+                20, DecisionSignalKind.EspPhaseChanged, T0.AddMinutes(17),
+                new Dictionary<string, string> { [SignalPayloadKeys.EspPhase] = "AccountSetup" })).NewState;
+            Assert.NotNull(state.AccountSetupEnteredUtc);
+
+            var step = ApplyDeviceSetupTerminalFailure(engine, state);
+
+            var advisory = SingleTimelineEffect(step, "esp_failure_advisory");
+            Assert.Equal("esp_failure_defanged_continueanyway_with_accountsetup", advisory.Parameters!["advisoryReason"]);
+            var deadline = FindDeadline(step.NewState, DeadlineNames.AdvisoryCompletion);
+            Assert.Equal(T0.AddMinutes(38).AddMinutes(30), deadline!.DueAtUtc);
+        }
+
+        // ========================================== real end: desktop completion ====
+
+        [Fact]
+        public void ObservationAdvisory_ThenRealUserDesktop_CompletesThroughFinalizing()
+        {
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = ApplyDeviceSetupTerminalFailure(engine, state).NewState;
+
+            // User pressed "Continue anyway" → signed in → DAD-validated real-user desktop.
+            var step = engine.Reduce(state, MakeSignal(
+                60, DecisionSignalKind.DesktopArrived, T0.AddMinutes(42)));
+
+            Assert.Equal(SessionStage.Finalizing, step.NewState.Stage);
+            Assert.Contains(":ContinueAnywayObservation", step.Transition.Trigger);
+            // Hello disabled → synthetic skip recorded by the fast-path.
+            Assert.NotNull(step.NewState.HelloResolvedUtc);
+            Assert.NotNull(FindDeadline(step.NewState, DeadlineNames.FinalizingGrace));
+        }
+
+        [Fact]
+        public void ObservationCompletion_EnrollmentComplete_CarriesEspSoftFailureMarker()
+        {
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = ApplyDeviceSetupTerminalFailure(engine, state).NewState;
+            state = engine.Reduce(state, MakeSignal(
+                60, DecisionSignalKind.DesktopArrived, T0.AddMinutes(42))).NewState;
+            Assert.Equal(SessionStage.Finalizing, state.Stage);
+
+            var step = engine.Reduce(state, DeadlineFired(70, T0.AddMinutes(42.1), DeadlineNames.FinalizingGrace));
+
+            Assert.Equal(SessionStage.Completed, step.NewState.Stage);
+            Assert.Equal(SessionOutcome.EnrollmentComplete, step.NewState.Outcome);
+
+            var complete = SingleTimelineEffect(step, "enrollment_complete");
+            var payload = Assert.IsType<Dictionary<string, object>>(complete.TypedPayload);
+            Assert.Equal("true", payload["espSoftFailure"]);
+            Assert.Equal("continue_anyway_observation", payload["completionSource"]);
+            Assert.Equal("DeviceSetup", payload["espSoftFailureCategory"]);
+        }
+
+        [Fact]
+        public void CleanCompletion_EnrollmentComplete_OmitsEspSoftFailureMarker()
+        {
+            // A session without any advisory must not carry the marker — regression guard for
+            // the BuildEnrollmentCompleteEffect change.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = engine.Reduce(state, MakeSignal(
+                20, DecisionSignalKind.EspPhaseChanged, T0.AddMinutes(17),
+                new Dictionary<string, string> { [SignalPayloadKeys.EspPhase] = "AccountSetup" })).NewState;
+            state = engine.Reduce(state, MakeSignal(
+                30, DecisionSignalKind.AccountSetupProvisioningComplete, T0.AddMinutes(20))).NewState;
+            state = engine.Reduce(state, MakeSignal(
+                40, DecisionSignalKind.DesktopArrived, T0.AddMinutes(21))).NewState;
+            Assert.Equal(SessionStage.Finalizing, state.Stage);
+
+            var step = engine.Reduce(state, DeadlineFired(50, T0.AddMinutes(21.1), DeadlineNames.FinalizingGrace));
+
+            Assert.Equal(SessionStage.Completed, step.NewState.Stage);
+            var complete = SingleTimelineEffect(step, "enrollment_complete");
+            var payload = Assert.IsType<Dictionary<string, object>>(complete.TypedPayload);
+            Assert.False(payload.ContainsKey("espSoftFailure"));
+            Assert.False(payload.ContainsKey("completionSource"));
+        }
+
+        // =============================================== window expiry / deadline ====
+
+        [Fact]
+        public void ObservationWindow_ExpiresWithoutDesktop_UnDefangsToEspTerminalFailure()
+        {
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = ApplyDeviceSetupTerminalFailure(engine, state).NewState;
+            Assert.Null(state.DesktopArrivedUtc);
+
+            var step = engine.Reduce(state, DeadlineFired(60, T0.AddMinutes(98), DeadlineNames.AdvisoryCompletion));
+
+            Assert.Equal(SessionStage.Failed, step.NewState.Stage);
+            Assert.Equal(SessionOutcome.EnrollmentFailed, step.NewState.Outcome);
+            // The failure cause IS the original ESP failure — same reason literal, likely-stuck
+            // promotion stays armed via LastFailureTrigger=EspTerminalFailure.
+            Assert.Equal(nameof(DecisionSignalKind.EspTerminalFailure), step.NewState.LastFailureTrigger!.Value);
+
+            var failed = SingleTimelineEffect(step, "enrollment_failed");
+            Assert.Equal("esp_terminal_failure", failed.Parameters!["reason"]);
+            Assert.Equal(
+                "continue_anyway_observation_window_expired_without_completion_evidence",
+                failed.Parameters!["advisoryReason"]);
+        }
+
+        [Fact]
+        public void ObservationWindow_FiresWithDesktopAndDisabledHello_CompletesInsteadOfFailing()
+        {
+            // Ordering variant: desktop arrived but (hypothetically) no eager completion ran —
+            // the deadline-fire conjunction must accept desktop + hello for the observation
+            // variant without the (structurally impossible) IME user-session gate.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine, helloPolicyDisabled: false);
+            state = ApplyDeviceSetupTerminalFailure(engine, state).NewState;
+            // Hello policy lands AFTER the failure; desktop arrives; no eager completion since
+            // the fast-path needs the policy fact at desktop-arrival time.
+            state = engine.Reduce(state, MakeSignal(
+                60, DecisionSignalKind.DesktopArrived, T0.AddMinutes(42))).NewState;
+            Assert.NotEqual(SessionStage.Finalizing, state.Stage);
+            state = engine.Reduce(state, MakeSignal(
+                65, DecisionSignalKind.HelloPolicyDetected, T0.AddMinutes(43),
+                new Dictionary<string, string> { [SignalPayloadKeys.HelloEnabled] = "false" })).NewState;
+
+            var step = engine.Reduce(state, DeadlineFired(70, T0.AddMinutes(98), DeadlineNames.AdvisoryCompletion));
+
+            Assert.Equal(SessionStage.Finalizing, step.NewState.Stage);
+            Assert.Null(step.NewState.Outcome);
+            Assert.NotNull(FindDeadline(step.NewState, DeadlineNames.FinalizingGrace));
+        }
+
+        [Fact]
+        public void ObservationWindow_FiresWithDesktopButUnknownHello_PromotesToAwaitingHello()
+        {
+            // Hello policy never observed (no PassportForWork value) — the hello prerequisite is
+            // structurally unresolvable. The never-observed promote must apply to the observation
+            // variant instead of failing a desktop-proven session.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine, helloPolicyDisabled: false);
+            state = ApplyDeviceSetupTerminalFailure(engine, state).NewState;
+            state = engine.Reduce(state, MakeSignal(
+                60, DecisionSignalKind.DesktopArrived, T0.AddMinutes(42))).NewState;
+            Assert.Null(state.HelloPolicyEnabled);
+
+            var step = engine.Reduce(state, DeadlineFired(70, T0.AddMinutes(98), DeadlineNames.AdvisoryCompletion));
+
+            Assert.Equal(SessionStage.AwaitingHello, step.NewState.Stage);
+            Assert.Null(step.NewState.Outcome);
+            Assert.NotNull(FindDeadline(step.NewState, DeadlineNames.HelloSafety));
+        }
+
+        // ======================================================= category recovery ====
+
+        [Fact]
+        public void ObservationAdvisory_DeviceSetupRecovery_ResolvesAdvisory()
+        {
+            // "Try again" path: the ESP re-runs the device phase and DeviceSetup resolves to
+            // success — the 4910a5a5 recovery hook must match the observation advisory too.
+            var engine = new DecisionEngine();
+            var state = SetupObservationEligibleSession(engine);
+            state = ApplyDeviceSetupTerminalFailure(engine, state).NewState;
+
+            var step = engine.Reduce(state, MakeSignal(
+                60, DecisionSignalKind.DeviceSetupProvisioningComplete, T0.AddMinutes(55)));
+
+            Assert.NotNull(step.NewState.EspAdvisoryFailureResolvedUtc);
+            Assert.Contains(step.Effects, e =>
+                e.Parameters != null
+                && e.Parameters.TryGetValue("eventType", out var et)
+                && et == "esp_failure_advisory_resolved");
+        }
+    }
+}

src/Backend/AutopilotMonitor.Functions.Tests/GatherRuleAuthorStampingTests.cs

@@ -0,0 +1,51 @@
+using System.Security.Claims;
+using AutopilotMonitor.Functions.Helpers;
+
+namespace AutopilotMonitor.Functions.Tests;
+
+/// <summary>
+/// Pins the claim precedence behind gather-rule Author stamping
+/// (<see cref="TenantHelper.GetUserDisplayName(ClaimsPrincipal)"/>): display name
+/// first, then UPN/email/preferred_username; null for anonymous/app-only callers
+/// so CreateGatherRule falls back to the model default "Autopilot Monitor".
+/// </summary>
+public class GatherRuleAuthorStampingTests
+{
+    private static ClaimsPrincipal Principal(params (string Type, string Value)[] claims)
+        => new(new ClaimsIdentity(claims.Select(c => new Claim(c.Type, c.Value)), "TestAuth"));
+
+    [Fact]
+    public void DisplayName_wins_over_upn()
+    {
+        var user = Principal(("name", "Alice Admin"), ("upn", "alice@contoso.com"));
+        Assert.Equal("Alice Admin", TenantHelper.GetUserDisplayName(user));
+    }
+
+    [Fact]
+    public void Upn_used_when_no_name_claim()
+    {
+        var user = Principal(("upn", "alice@contoso.com"), ("preferred_username", "alice@contoso.onmicrosoft.com"));
+        Assert.Equal("alice@contoso.com", TenantHelper.GetUserDisplayName(user));
+    }
+
+    [Fact]
+    public void PreferredUsername_is_the_last_fallback()
+    {
+        var user = Principal(("preferred_username", "alice@contoso.com"));
+        Assert.Equal("alice@contoso.com", TenantHelper.GetUserDisplayName(user));
+    }
+
+    [Fact]
+    public void Null_for_principal_without_identifying_claims()
+    {
+        Assert.Null(TenantHelper.GetUserDisplayName(Principal(("tid", "a1b2c3d4-e5f6-7890-abcd-ef1234567890"))));
+    }
+
+    [Fact]
+    public void Null_for_unauthenticated_or_missing_principal()
+    {
+        Assert.Null(TenantHelper.GetUserDisplayName((ClaimsPrincipal?)null));
+        // ClaimsIdentity without an authentication type => IsAuthenticated == false.
+        Assert.Null(TenantHelper.GetUserDisplayName(new ClaimsPrincipal(new ClaimsIdentity(new[] { new Claim("name", "Ghost") }))));
+    }
+}

src/Backend/AutopilotMonitor.Functions.Tests/GatherRuleUpdatePartialMergeTests.cs

@@ -33,6 +33,7 @@ public class GatherRuleUpdatePartialMergeTests
         Enabled = true,
         IsBuiltIn = false,
         IsCommunity = false,
+        Author = "Alice Admin",
         CreatedAt = new DateTime(2026, 1, 15, 12, 0, 0, DateTimeKind.Utc),
         ActiveFromPhase = "AccountSetup",
         EmitMode = "on_change",
@@ -77,6 +78,7 @@ public class GatherRuleUpdatePartialMergeTests
         Assert.Equal("AccountSetup", written.ActiveFromPhase);
         Assert.Equal("on_change", written.EmitMode);
         Assert.Equal(existing.CreatedAt, written.CreatedAt);
+        Assert.Equal("Alice Admin", written.Author);
     }
 
     [Fact]
@@ -126,6 +128,33 @@ public class GatherRuleUpdatePartialMergeTests
         Assert.Equal(existing.CreatedAt, written.CreatedAt);
     }
 
+    [Fact]
+    public async Task FullPayload_preserves_original_Author_even_when_payload_carries_one()
+    {
+        var existing = ExistingCustomRule();
+        var (service, _, stored) = BuildService(existing);
+
+        var fullPayload = new GatherRule
+        {
+            RuleId = RuleId,
+            Title = "Collect BIOS Config v2",
+            CollectorType = "registry",
+            Target = "HKLM\\SOFTWARE\\RealmJoin\\Custom\\BIOS2",
+            Trigger = "interval",
+            OutputEventType = "gather_bios_config",
+            Enabled = true,
+            // A hand-crafted PUT could try to rewrite attribution — Author is
+            // stamped at create time and immutable afterwards.
+            Author = "Spoofed Author",
+        };
+
+        var ok = await service.UpdateRuleAsync(TenantId, fullPayload);
+
+        Assert.True(ok);
+        var written = Assert.Single(stored);
+        Assert.Equal("Alice Admin", written.Author);
+    }
+
     [Fact]
     public async Task BuiltInRule_toggle_still_goes_through_RuleState_path()
     {

src/Backend/AutopilotMonitor.Functions.Tests/Offboarding/TenantCustomsArchiveContractTests.cs

@@ -42,6 +42,7 @@ public sealed class TenantCustomsArchiveContractTests
             reEnqueuer: null!,
             opsEvents: null!,
             customsArchive: archive,
+            configRepo: null!,
             farewellEmail: new FakeOffboardFarewellEmailSender(),
             logger: NullLogger<TenantOffboardingHandler>.Instance);
     }

src/Backend/AutopilotMonitor.Functions.Tests/Offboarding/TenantOffboardingHandlerDrainContractTests.cs

@@ -9,6 +9,7 @@ using Azure;
 using Azure.Data.Tables;
 using Azure.Storage.Blobs;
 using AutopilotMonitor.Functions.DataAccess.TableStorage;
+using AutopilotMonitor.Functions.Functions.Admin;
 using AutopilotMonitor.Functions.Services;
 using AutopilotMonitor.Functions.Services.Deletion;
 using AutopilotMonitor.Functions.Services.Notifications;
@@ -301,6 +302,47 @@ public class TenantOffboardingHandlerDrainContractTests
             "Phase 2.G must delete the Expectations blob as its last step");
     }
 
+    [Fact]
+    public async Task UnoffboardRace_ConfigNoLongerTombstone_FailsClosed_NoSafeWipe()
+    {
+        // 2.D-guard: an operator re-enabled the tenant while the cascade was queued/draining
+        // (the config write paths refuse this, but the guard is the last line of defense).
+        // Wiping now would destroy data of a live tenant → fail-closed, zero wipes.
+        var harness = Harness.New();
+        harness.ConfigRepo
+            .Setup(r => r.GetTenantConfigurationAsync(TenantId))
+            .ReturnsAsync(new TenantConfiguration { TenantId = TenantId, Disabled = false });
+
+        await Assert.ThrowsAsync<InvalidOperationException>(() =>
+            harness.Sut.HandleAsync(harness.Envelope(drainPollCount: 0)));
+
+        Assert.Equal("Failed", harness.Repo.Markers[TenantId].Status);
+        Assert.Equal("unoffboard_race", harness.Repo.Markers[TenantId].FailedPhase);
+        Assert.Equal("Failed", harness.Repo.History[HistoryRowKey].Status);
+        Assert.Equal(0, harness.SafeWipeProbe.WipeCallCount);
+    }
+
+    [Fact]
+    public async Task UnoffboardGuard_TombstoneStillInPlace_ProceedsToWipeAndCompleted()
+    {
+        // The normal case: Phase 1's tombstone is untouched → the guard must NOT get in
+        // the way of a legitimate offboarding.
+        var harness = Harness.New();
+        harness.ConfigRepo
+            .Setup(r => r.GetTenantConfigurationAsync(TenantId))
+            .ReturnsAsync(new TenantConfiguration
+            {
+                TenantId = TenantId,
+                Disabled = true,
+                DisabledReason = TenantOffboardFunction.OffboardingDisabledReason,
+            });
+
+        await harness.Sut.HandleAsync(harness.Envelope(drainPollCount: 0));
+
+        Assert.Equal("Completed", harness.Repo.Markers[TenantId].Status);
+        Assert.True(harness.SafeWipeProbe.WipeCallCount > 0);
+    }
+
     [Fact]
     public async Task SessionNotFound_CountsAsSatisfied_NoDrainProbeCall()
     {
@@ -393,6 +435,7 @@ public class TenantOffboardingHandlerDrainContractTests
         public Mock<OffboardingSessionEnumerator> Enumerator { get; }
         public CountingSafeWipeService SafeWipeProbe { get; } = new();
         public Mock<IMaintenanceRepository> Maintenance { get; } = new();
+        public Mock<IConfigRepository> ConfigRepo { get; } = new();
         public List<string> EnumeratorYields { get; } = new();
         public Exception? EnumeratorThrow { get; set; }
 
@@ -450,6 +493,7 @@ public class TenantOffboardingHandlerDrainContractTests
                 h.ReEnqueuer.Object,
                 opsService,
                 Mock.Of<ITenantCustomsArchiveRepository>(),
+                h.ConfigRepo.Object,
                 new FakeOffboardFarewellEmailSender(),
                 NullLogger<TenantOffboardingHandler>.Instance);
 
@@ -622,10 +666,11 @@ public class TenantOffboardingHandlerDrainContractTests
             ITenantOffboardingEnqueuer reEnqueuer,
             OpsEventService opsEvents,
             ITenantCustomsArchiveRepository customsArchive,
+            IConfigRepository configRepo,
             IOffboardFarewellEmailSender farewellEmail,
             ILogger<TenantOffboardingHandler> logger)
             : base(auditRepo, enumerator, cascadeEnqueuer, expectations, drainProbe, safeWipe,
-                   storage, maintenance, reEnqueuer, opsEvents, customsArchive, farewellEmail, logger)
+                   storage, maintenance, reEnqueuer, opsEvents, customsArchive, configRepo, farewellEmail, logger)
         {
         }
 

src/Backend/AutopilotMonitor.Functions.Tests/Offboarding/TenantOffboardingHandlerOrderingTests.cs

@@ -358,6 +358,7 @@ public class TenantOffboardingHandlerOrderingTests
                 h.ReEnqueuer.Object,
                 BuildOpsService(),
                 Mock.Of<ITenantCustomsArchiveRepository>(),
+                Mock.Of<IConfigRepository>(),
                 h.FarewellEmail,
                 NullLogger<TenantOffboardingHandler>.Instance);
 
@@ -459,9 +460,9 @@ public class TenantOffboardingHandlerOrderingTests
             IOffboardingAuditRepository a, OffboardingSessionEnumerator e, ISessionDeletionEnqueuer c,
             IOffboardingExpectationsStore exp, IDeletionProgressDrainProbe d, SafeWipeService sw,
             TableStorageService s, IMaintenanceRepository m, ITenantOffboardingEnqueuer re,
-            OpsEventService o, ITenantCustomsArchiveRepository ca, IOffboardFarewellEmailSender fe,
-            ILogger<TenantOffboardingHandler> log)
-            : base(a, e, c, exp, d, sw, s, m, re, o, ca, fe, log) { }
+            OpsEventService o, ITenantCustomsArchiveRepository ca, IConfigRepository cr,
+            IOffboardFarewellEmailSender fe, ILogger<TenantOffboardingHandler> log)
+            : base(a, e, c, exp, d, sw, s, m, re, o, ca, cr, fe, log) { }
 
         internal override Task ArchiveAndWipeRulesTableAsync(
             string tableName, string tenantId, string historyRowKey, CancellationToken ct)

src/Backend/AutopilotMonitor.Functions.Tests/Offboarding/TenantOffboardingWorkerPoisonTests.cs

@@ -245,6 +245,7 @@ public class TenantOffboardingWorkerPoisonTests
                 Mock.Of<ITenantOffboardingEnqueuer>(),
                 BuildOpsService(),
                 Mock.Of<ITenantCustomsArchiveRepository>(),
+                Mock.Of<IConfigRepository>(),
                 new FakeOffboardFarewellEmailSender(),
                 NullLogger<TenantOffboardingHandler>.Instance);
         }
@@ -255,9 +256,9 @@ public class TenantOffboardingWorkerPoisonTests
                 IOffboardingAuditRepository a, OffboardingSessionEnumerator e, ISessionDeletionEnqueuer c,
                 IOffboardingExpectationsStore exp, IDeletionProgressDrainProbe d, SafeWipeService sw,
                 TableStorageService s, IMaintenanceRepository m, ITenantOffboardingEnqueuer re,
-                OpsEventService o, ITenantCustomsArchiveRepository ca, IOffboardFarewellEmailSender fe,
-                ILogger<TenantOffboardingHandler> log)
-                : base(a, e, c, exp, d, sw, s, m, re, o, ca, fe, log) { }
+                OpsEventService o, ITenantCustomsArchiveRepository ca, IConfigRepository cr,
+                IOffboardFarewellEmailSender fe, ILogger<TenantOffboardingHandler> log)
+                : base(a, e, c, exp, d, sw, s, m, re, o, ca, cr, fe, log) { }
 
             internal override Task ArchiveAndWipeRulesTableAsync(
                 string tableName, string tenantId, string historyRowKey, CancellationToken ct)

src/Backend/AutopilotMonitor.Functions.Tests/TenantConfigPatchServiceTests.cs

@@ -4,6 +4,7 @@ using System.Linq;
 using System.Threading;
 using System.Threading.Tasks;
 using AutopilotMonitor.Functions.DataAccess.TableStorage;
+using AutopilotMonitor.Functions.Functions.Admin;
 using AutopilotMonitor.Functions.Services;
 using AutopilotMonitor.Shared;
 using AutopilotMonitor.Shared.DataAccess;
@@ -154,6 +155,68 @@ public class TenantConfigPatchServiceTests
         Assert.Null(harness.Current!.ContactEmail);
     }
 
+    // ── Mid-offboarding freeze ──────────────────────────────────────────────
+
+    [Fact]
+    public async Task Patch_OffboardingTombstone_IsFrozen_EvenForGa()
+    {
+        // While the row is the offboarding tombstone, the cascade owns it: a GA patch
+        // lifting Disabled would un-gate agents against a tenant whose partitions are
+        // queued for deletion (un-offboard race).
+        var stored = Stored();
+        stored.Disabled = true;
+        stored.DisabledReason = TenantOffboardFunction.OffboardingDisabledReason;
+        var harness = new Harness(stored);
+
+        var outcome = await harness.Sut.ApplyFieldPatchAsync(
+            TenantId, Fields(("disabled", false)), Ga, "mcp-patch", "trying to un-offboard");
+
+        Assert.False(outcome.Success);
+        Assert.Equal(PatchFailure.ValidationFailed, outcome.Failure);
+        Assert.Contains("offboarding", outcome.Error, StringComparison.OrdinalIgnoreCase);
+        Assert.True(harness.Current!.Disabled, "tombstone must be untouched");
+        Assert.Equal(0, harness.ReplaceCalls);
+    }
+
+    [Fact]
+    public async Task Patch_OrdinarySuspension_IsNotFrozen()
+    {
+        // The freeze is specific to the offboarding tombstone — a normally suspended
+        // tenant (different DisabledReason) stays fully patchable.
+        var stored = Stored();
+        stored.Disabled = true;
+        stored.DisabledReason = "Manual suspension";
+        var harness = new Harness(stored);
+
+        var outcome = await harness.Sut.ApplyFieldPatchAsync(
+            TenantId, Fields(("dataRetentionDays", 60)), Ga, "mcp-patch", null);
+
+        Assert.True(outcome.Success);
+        Assert.Equal(60, harness.Current!.DataRetentionDays);
+    }
+
+    [Fact]
+    public async Task Revert_OffboardingTombstone_IsFrozen()
+    {
+        // A revert restores a pre-offboarding snapshot with Disabled=false — the exact
+        // un-offboard race. Build a backup while the tenant is healthy, then tombstone.
+        var harness = new Harness(Stored());
+        var patched = await harness.Sut.ApplyFieldPatchAsync(
+            TenantId, Fields(("dataRetentionDays", 90)), Ga, "mcp-patch", null);
+        Assert.True(patched.Success);
+
+        harness.Current!.Disabled = true;
+        harness.Current.DisabledReason = TenantOffboardFunction.OffboardingDisabledReason;
+
+        var outcome = await harness.Sut.RevertAsync(
+            TenantId, backupId: null, includeProtectedFields: false, Ga, "mcp-revert", null);
+
+        Assert.False(outcome.Success);
+        Assert.Equal(PatchFailure.ValidationFailed, outcome.Failure);
+        Assert.Contains("offboarding", outcome.Error, StringComparison.OrdinalIgnoreCase);
+        Assert.True(harness.Current.Disabled, "tombstone must be untouched");
+    }
+
     [Fact]
     public async Task Patch_NoOp_ValueEqualsStored_NoBackupNoWrite()
     {

src/Backend/AutopilotMonitor.Functions/DataAccess/TableStorage/TableConfigRepository.cs

@@ -623,6 +623,7 @@ namespace AutopilotMonitor.Functions.DataAccess.TableStorage
                 { "NtpServer", config.NtpServer },
                 { "EnableImeMatchLog", config.EnableImeMatchLog },
                 { "EnableGatherRuleDebugLog", config.EnableGatherRuleDebugLog },
+                { "EnableEspContinueAnywayObservation", config.EnableEspContinueAnywayObservation },
                 { "LogLevel", config.LogLevel },
                 { "MaxBatchSize", config.MaxBatchSize },
                 { "DiagnosticsBlobSasUrl", config.DiagnosticsBlobSasUrl },
@@ -729,6 +730,7 @@ namespace AutopilotMonitor.Functions.DataAccess.TableStorage
                 NtpServer = string.IsNullOrWhiteSpace(entity.GetString("NtpServer")) ? "time.windows.com" : entity.GetString("NtpServer"),
                 EnableImeMatchLog = entity.GetBoolean("EnableImeMatchLog"),
                 EnableGatherRuleDebugLog = entity.GetBoolean("EnableGatherRuleDebugLog"),
+                EnableEspContinueAnywayObservation = entity.GetBoolean("EnableEspContinueAnywayObservation"),
                 LogLevel = entity.GetString("LogLevel"),
                 MaxBatchSize = entity.GetInt32("MaxBatchSize"),
                 DiagnosticsBlobSasUrl = entity.GetString("DiagnosticsBlobSasUrl"),

src/Backend/AutopilotMonitor.Functions/DataAccess/TableStorage/TableSessionRepository.cs

@@ -65,13 +65,14 @@ namespace AutopilotMonitor.Functions.DataAccess.TableStorage
             bool? isUserDriven = null, DateTime? resumedAt = null,
             DateTime? stalledAt = null, bool clearStalledAt = false, bool clearFailureReason = false,
             string? failureSource = null, string? adminMarkedAction = null,
-            string? failureSnapshotJson = null, bool allowTerminalReclassification = false)
+            string? failureSnapshotJson = null, bool allowTerminalReclassification = false,
+            bool espSoftFailure = false, string? completionSource = null)
         {
             var transitioned = await _storage.UpdateSessionStatusAsync(tenantId, sessionId, status,
                 currentPhase, failureReason, completedAt, earliestEventTimestamp,
                 latestEventTimestamp, isPreProvisioned, isUserDriven, resumedAt,
                 stalledAt, clearStalledAt, clearFailureReason, failureSource, adminMarkedAction,
-                failureSnapshotJson, allowTerminalReclassification);
+                failureSnapshotJson, allowTerminalReclassification, espSoftFailure, completionSource);
 
             // Reconcile the authoritative EventCount + RebootCount whenever a session actually
             // transitions to a terminal state through ANY caller — ingest, admin mark

src/Backend/AutopilotMonitor.Functions/Functions/Admin/TenantOffboardFunction.cs

@@ -517,6 +517,17 @@ public class TenantOffboardFunction
     /// </summary>
     internal const string OffboardingDisabledReason = "Offboarding in progress";
 
+    /// <summary>
+    /// True when the row is the offboarding tombstone written by
+    /// <c>EnsureTenantDisabledAsync</c>. Shared predicate for every config writer and
+    /// wiper that must distinguish the mid-offboarding freeze from an ordinary
+    /// suspended (or re-onboarded) tenant: while this is true, the tenant's
+    /// configuration is frozen and the offboarding cascade owns the row.
+    /// </summary>
+    internal static bool IsOffboardingTombstone(TenantConfiguration? config) =>
+        config is { Disabled: true }
+        && string.Equals(config.DisabledReason, OffboardingDisabledReason, StringComparison.Ordinal);
+
     /// <summary>
     /// Result of <see cref="EnsureTenantDisabledAsync"/>. Carries both whether a fresh
     /// write happened AND the tombstone's <c>LastUpdated</c> timestamp so the resume-path

src/Backend/AutopilotMonitor.Functions/Functions/Config/GetAgentConfigFunction.cs

@@ -274,13 +274,14 @@ namespace AutopilotMonitor.Functions.Functions.Config
             var response = req.CreateResponse(HttpStatusCode.OK);
             await response.WriteAsJsonAsync(new AgentConfigResponse
             {
-                ConfigVersion = 36, // EnableGatherRuleDebugLog (local gather-rule evaluation trace)
+                ConfigVersion = 37, // EnableEspContinueAnywayObservation (Continue-Anyway observation mode)
                 UploadIntervalSeconds = 10,
                 SelfDestructOnComplete = tenantConfig.SelfDestructOnComplete ?? true,
                 KeepLogFile = tenantConfig.KeepLogFile ?? false,
                 EnableGeoLocation = tenantConfig.EnableGeoLocation ?? true,
                 EnableImeMatchLog = tenantConfig.EnableImeMatchLog ?? false,
                 EnableGatherRuleDebugLog = tenantConfig.EnableGatherRuleDebugLog ?? false,
+                EnableEspContinueAnywayObservation = tenantConfig.EnableEspContinueAnywayObservation ?? false,
                 MaxAuthFailures = tenantConfig.MaxAuthFailures ?? 5,
                 AuthFailureTimeoutMinutes = tenantConfig.AuthFailureTimeoutMinutes ?? 0,
                 LogLevel = tenantConfig.LogLevel ?? "Info",

src/Backend/AutopilotMonitor.Functions/Functions/Config/UpdateTenantConfigurationFunction.cs

@@ -3,6 +3,7 @@ using System.IO;
 using System.Net;
 using System.Threading.Tasks;
 using System.Collections.Generic;
+using AutopilotMonitor.Functions.Functions.Admin;
 using AutopilotMonitor.Functions.Helpers;
 using AutopilotMonitor.Functions.Security;
 using AutopilotMonitor.Functions.Services;
@@ -71,6 +72,23 @@ namespace AutopilotMonitor.Functions.Functions.Config
                 // before validation/save and (2) protect GA-only fields below.
                 var existingConfig = await _configService.GetConfigurationAsync(requestCtx.TargetTenantId);
 
+                // Mid-offboarding freeze: the Disabled tombstone written by TenantOffboardFunction
+                // is the cascade's ONLY auth gate. A PUT that lifts it (or edits anything else)
+                // races the queued wipe — data written afterwards gets destroyed, and agents
+                // would be un-gated against a tenant whose partitions are being deleted. Refuse
+                // for ALL callers, GA included; once the cascade completes the row is deleted
+                // and re-onboarding starts from a fresh default config anyway.
+                if (TenantOffboardFunction.IsOffboardingTombstone(existingConfig))
+                {
+                    var conflict = req.CreateResponse(HttpStatusCode.Conflict);
+                    await conflict.WriteAsJsonAsync(new
+                    {
+                        success = false,
+                        message = "Tenant offboarding is in progress — the configuration is frozen until the cascade completes."
+                    });
+                    return conflict;
+                }
+
                 // Defense-in-depth: a read-only GlobalReader is served a redacted config (secrets replaced
                 // with the ***REDACTED*** sentinel). If such a view is ever round-tripped back on a save,
                 // never persist the placeholder — restore the real secret from the stored config. (Own-tenant
@@ -101,6 +119,7 @@ namespace AutopilotMonitor.Functions.Functions.Config
                         config.BootstrapTokenEnabled != existingConfig.BootstrapTokenEnabled ||
                         config.UnrestrictedModeEnabled != existingConfig.UnrestrictedModeEnabled ||
                         config.EntraAppRolesEnabled != existingConfig.EntraAppRolesEnabled ||
+                        config.EnableEspContinueAnywayObservation != existingConfig.EnableEspContinueAnywayObservation ||
                         config.CustomRateLimitRequestsPerMinute != existingConfig.CustomRateLimitRequestsPerMinute ||
                         config.CustomUserRateLimitRequestsPerMinute != existingConfig.CustomUserRateLimitRequestsPerMinute ||
                         config.Disabled != existingConfig.Disabled ||
@@ -115,6 +134,9 @@ namespace AutopilotMonitor.Functions.Functions.Config
                     config.BootstrapTokenEnabled = existingConfig.BootstrapTokenEnabled;
                     config.UnrestrictedModeEnabled = existingConfig.UnrestrictedModeEnabled;
                     config.EntraAppRolesEnabled = existingConfig.EntraAppRolesEnabled;
+                    // Continue-Anyway observation is an operator-set behavioral override —
+                    // a tenant admin must not be able to relax their own failure semantics.
+                    config.EnableEspContinueAnywayObservation = existingConfig.EnableEspContinueAnywayObservation;
                     config.CustomRateLimitRequestsPerMinute = existingConfig.CustomRateLimitRequestsPerMinute;
                     config.CustomUserRateLimitRequestsPerMinute = existingConfig.CustomUserRateLimitRequestsPerMinute;
                     config.Disabled = existingConfig.Disabled;

src/Backend/AutopilotMonitor.Functions/Functions/Rules/GatherRulesFunction.cs

@@ -72,6 +72,10 @@ namespace AutopilotMonitor.Functions.Functions.Rules
                 return badRequest;
             }
 
+            // Author is stamped from the creator's token, never from the payload
+            // (anti-spoof), and stays immutable through every later update.
+            rule.Author = TenantHelper.GetUserDisplayName(req) ?? "Autopilot Monitor";
+
             try
             {
                 var success = await _ruleService.CreateRuleAsync(tenantId, rule);

src/Backend/AutopilotMonitor.Functions/Helpers/TenantHelper.cs

@@ -19,27 +19,33 @@ public static class TenantHelper
     /// <param name="req">The HTTP request</param>
     /// <returns>The tenant ID from the JWT token</returns>
     /// <exception cref="UnauthorizedAccessException">Thrown when user is not authenticated or tenant ID is missing</exception>
-    public static string GetTenantId(HttpRequestData req)
+    /// <summary>
+    /// Resolves the authenticated ClaimsPrincipal for the request, or null.
+    /// Azure Functions Isolated Worker: FunctionContext.Items first (set by
+    /// AuthenticationMiddleware, more reliable than httpContext.User), then
+    /// the HTTP context as fallback.
+    /// </summary>
+    private static ClaimsPrincipal? GetAuthenticatedPrincipal(HttpRequestData req)
     {
-        // Azure Functions Isolated Worker: Try FunctionContext.Items first
-        // This is set by AuthenticationMiddleware and is more reliable than httpContext.User
-        ClaimsPrincipal? user = null;
-
         if (req.FunctionContext.Items.TryGetValue("ClaimsPrincipal", out var principalObj)
             && principalObj is ClaimsPrincipal principal)
         {
-            user = principal;
+            return principal;
         }
-        else
+
+        var httpContext = req.FunctionContext.GetHttpContext();
+        if (httpContext?.User?.Identity?.IsAuthenticated == true)
         {
-            // Fallback to HTTP context (may not work reliably in isolated worker)
-            var httpContext = req.FunctionContext.GetHttpContext();
-            if (httpContext?.User?.Identity?.IsAuthenticated == true)
-            {
-                user = httpContext.User;
-            }
+            return httpContext.User;
         }
 
+        return null;
+    }
+
+    public static string GetTenantId(HttpRequestData req)
+    {
+        var user = GetAuthenticatedPrincipal(req);
+
         if (user?.Identity?.IsAuthenticated != true)
         {
             throw new UnauthorizedAccessException("User is not authenticated. JWT token required.");
@@ -67,23 +73,7 @@ public static class TenantHelper
     /// <returns>User email or name, or "Anonymous" if not authenticated</returns>
     public static string GetUserIdentifier(HttpRequestData req)
     {
-        // Azure Functions Isolated Worker: Try FunctionContext.Items first
-        ClaimsPrincipal? user = null;
-
-        if (req.FunctionContext.Items.TryGetValue("ClaimsPrincipal", out var principalObj)
-            && principalObj is ClaimsPrincipal principal)
-        {
-            user = principal;
-        }
-        else
-        {
-            // Fallback to HTTP context
-            var httpContext = req.FunctionContext.GetHttpContext();
-            if (httpContext?.User?.Identity?.IsAuthenticated == true)
-            {
-                user = httpContext.User;
-            }
-        }
+        var user = GetAuthenticatedPrincipal(req);
 
         if (user?.Identity?.IsAuthenticated != true)
         {
@@ -100,4 +90,32 @@ public static class TenantHelper
                user.FindFirst("name")?.Value ??
                "Unknown";
     }
+
+    /// <summary>
+    /// Gets the caller's human-readable name for attribution (e.g. rule Author):
+    /// display name first, then UPN/email/preferred_username. Returns null when
+    /// the request carries no user identity (anonymous or app-only token) so the
+    /// caller can choose its own fallback.
+    /// </summary>
+    public static string? GetUserDisplayName(HttpRequestData req)
+        => GetUserDisplayName(GetAuthenticatedPrincipal(req));
+
+    /// <summary>
+    /// Claim-precedence core of <see cref="GetUserDisplayName(HttpRequestData)"/>,
+    /// separated so the precedence is unit-testable without an HttpRequestData.
+    /// </summary>
+    public static string? GetUserDisplayName(ClaimsPrincipal? user)
+    {
+        if (user?.Identity?.IsAuthenticated != true)
+        {
+            return null;
+        }
+
+        return user.FindFirst("name")?.Value ??
+               user.FindFirst(ClaimTypes.Name)?.Value ??
+               user.FindFirst("upn")?.Value ??
+               user.FindFirst(ClaimTypes.Upn)?.Value ??
+               user.FindFirst(ClaimTypes.Email)?.Value ??
+               user.FindFirst("preferred_username")?.Value;
+    }
 }

src/Backend/AutopilotMonitor.Functions/Services/Analyze/AzureQueueAnalyzeOnEnrollmentEndProducer.cs

@@ -28,6 +28,19 @@ namespace AutopilotMonitor.Functions.Services.Analyze
 
         private int _queueEnsured; // 0 = not yet ensured, 1 = CreateIfNotExistsAsync has run
 
+        /// <summary>
+        /// Delay before the enqueued message becomes visible to the worker. The trigger event
+        /// (enrollment_complete / enrollment_failed) is by design the FIRST event of the
+        /// agent's terminal sequence — the terminal flush (stuck-app promotion's synthetic
+        /// app_install_failed, app_tracking_summary, esp_apps_failure_correlation,
+        /// agent_shutting_down) lands in separate uploads seconds later. An immediately
+        /// visible message let the worker evaluate rules against a stream missing those rows
+        /// (session 53d1e9f6: ANALYZE-ESP-004's optional app_install_failed condition never
+        /// matched, {{appName}} stayed a literal placeholder — and the RuleResult dedup makes
+        /// that permanent). 30 s covers the flush including the agent's late-event grace.
+        /// </summary>
+        private static readonly TimeSpan s_visibilityDelay = TimeSpan.FromSeconds(30);
+
         public AzureQueueAnalyzeOnEnrollmentEndProducer(
             QueueClientFactory queueFactory,
             ILogger<AzureQueueAnalyzeOnEnrollmentEndProducer> logger)
@@ -53,7 +66,10 @@ namespace AutopilotMonitor.Functions.Services.Analyze
             try
             {
                 var body = JsonConvert.SerializeObject(envelope);
-                await _queueClient.SendMessageAsync(body, cancellationToken).ConfigureAwait(false);
+                await _queueClient.SendMessageAsync(
+                    body,
+                    visibilityTimeout: s_visibilityDelay,
+                    cancellationToken: cancellationToken).ConfigureAwait(false);
                 _logger.LogInformation(
                     "Analyze enqueued (tenant={Tenant} session={Session} reason={Reason})",
                     envelope.TenantId, envelope.SessionId, envelope.Reason);

src/Backend/AutopilotMonitor.Functions/Services/EmailTemplates.cs

@@ -39,9 +39,9 @@ public static class EmailTemplates
 
         <!-- Header -->
         <tr>
-          <td style=""background:linear-gradient(135deg,#2563eb,#4f46e5); padding:32px 40px; text-align:center;"">
+          <td style=""background:linear-gradient(135deg,#1e8a4c,#33b161); padding:32px 40px; text-align:center;"">
             <h1 style=""color:#ffffff; margin:0; font-size:24px; font-weight:700;"">Autopilot Monitor</h1>
-            <p style=""color:#c7d2fe; margin:8px 0 0; font-size:14px;"">Windows Autopilot monitoring</p>
+            <p style=""color:#c9efd8; margin:8px 0 0; font-size:14px;"">Windows Autopilot monitoring</p>
           </td>
         </tr>
 
@@ -68,8 +68,8 @@ public static class EmailTemplates
             </p>
 
             <ul style=""color:#374151; font-size:14px; line-height:1.8; margin:0 0 24px; padding-left:20px;"">
-              <li><a href=""https://github.com/okieselbach/AutopilotMonitor/issues"" target=""_blank"" style=""color:#2563eb; text-decoration:underline;"">Open a GitHub Issue</a></li>
-              <li><a href=""https://www.linkedin.com/in/oliver-kieselbach/"" target=""_blank"" style=""color:#2563eb; text-decoration:underline;"">Connect on LinkedIn</a></li>
+              <li><a href=""https://github.com/okieselbach/AutopilotMonitor/issues"" target=""_blank"" style=""color:#1e8a4c; text-decoration:underline;"">Open a GitHub Issue</a></li>
+              <li><a href=""https://www.linkedin.com/in/oliver-kieselbach/"" target=""_blank"" style=""color:#1e8a4c; text-decoration:underline;"">Connect on LinkedIn</a></li>
             </ul>
 
             <p style=""color:#374151; font-size:15px; line-height:1.6; margin:0 0 24px;"">
@@ -117,9 +117,9 @@ public static class EmailTemplates
 
         <!-- Header -->
         <tr>
-          <td style=""background:linear-gradient(135deg,#2563eb,#4f46e5); padding:32px 40px; text-align:center;"">
+          <td style=""background:linear-gradient(135deg,#1e8a4c,#33b161); padding:32px 40px; text-align:center;"">
             <h1 style=""color:#ffffff; margin:0; font-size:24px; font-weight:700;"">Autopilot Monitor</h1>
-            <p style=""color:#c7d2fe; margin:8px 0 0; font-size:14px;"">Windows Autopilot monitoring</p>
+            <p style=""color:#c9efd8; margin:8px 0 0; font-size:14px;"">Windows Autopilot monitoring</p>
           </td>
         </tr>
 
@@ -140,7 +140,7 @@ public static class EmailTemplates
             <!-- CTA Button -->
             <table cellpadding=""0"" cellspacing=""0"" style=""margin:0 auto 32px;"">
               <tr>
-                <td style=""background-color:#2563eb; border-radius:8px;"">
+                <td style=""background-color:#1e8a4c; border-radius:8px;"">
                   <a href=""{Constants.DocsBaseUrl}"" target=""_blank""
                      style=""display:inline-block; padding:14px 32px; color:#ffffff; font-size:15px; font-weight:600; text-decoration:none;"">
                     View Documentation
@@ -164,8 +164,8 @@ public static class EmailTemplates
             </p>
 
             <ul style=""color:#374151; font-size:14px; line-height:1.8; margin:0 0 24px; padding-left:20px;"">
-              <li><a href=""https://github.com/okieselbach/AutopilotMonitor/issues"" target=""_blank"" style=""color:#2563eb; text-decoration:underline;"">Open a GitHub Issue</a></li>
-              <li><a href=""https://www.linkedin.com/in/oliver-kieselbach/"" target=""_blank"" style=""color:#2563eb; text-decoration:underline;"">Connect on LinkedIn</a></li>
+              <li><a href=""https://github.com/okieselbach/AutopilotMonitor/issues"" target=""_blank"" style=""color:#1e8a4c; text-decoration:underline;"">Open a GitHub Issue</a></li>
+              <li><a href=""https://www.linkedin.com/in/oliver-kieselbach/"" target=""_blank"" style=""color:#1e8a4c; text-decoration:underline;"">Connect on LinkedIn</a></li>
             </ul>
 
             <p style=""color:#6b7280; font-size:14px; line-height:1.6; margin:0;"">

src/Backend/AutopilotMonitor.Functions/Services/EventIngestProcessor.Classification.cs

@@ -391,11 +391,24 @@ namespace AutopilotMonitor.Functions.Services
 
             if (c.CompletionEvent != null)
             {
+                // Soft-failure marker from the V2 engine's enrollment_complete audit payload:
+                // the session completed while an ESP-failure advisory was still unresolved
+                // (Continue-Anyway observation / classic advisory defang). Stamped on the
+                // Sessions row so the UI can render the amber "completed with issues" badge.
+                var espSoftFailure = string.Equals(
+                    c.CompletionEvent.Data?.ContainsKey("espSoftFailure") == true
+                        ? c.CompletionEvent.Data["espSoftFailure"]?.ToString() : null,
+                    "true", StringComparison.OrdinalIgnoreCase);
+                var completionSource = c.CompletionEvent.Data?.ContainsKey("completionSource") == true
+                    ? c.CompletionEvent.Data["completionSource"]?.ToString() : null;
+
                 statusTransitioned = await _sessionRepo.UpdateSessionStatusAsync(
                     request.TenantId, request.SessionId, SessionStatus.Succeeded, c.CompletionEvent.Phase,
                     completedAt: c.CompletionEvent.Timestamp,
-                    earliestEventTimestamp: c.EarliestEventTimestamp, latestEventTimestamp: c.LatestEventTimestamp);
-                _logger.LogInformation("{SessionPrefix} Status: Succeeded (transitioned={Transitioned})", sessionPrefix, statusTransitioned);
+                    earliestEventTimestamp: c.EarliestEventTimestamp, latestEventTimestamp: c.LatestEventTimestamp,
+                    espSoftFailure: espSoftFailure, completionSource: completionSource);
+                _logger.LogInformation("{SessionPrefix} Status: Succeeded (transitioned={Transitioned}, espSoftFailure={EspSoftFailure})",
+                    sessionPrefix, statusTransitioned, espSoftFailure);
             }
             else if (c.FailureEvent != null)
             {

src/Backend/AutopilotMonitor.Functions/Services/GatherRuleService.cs

@@ -214,7 +214,12 @@ namespace AutopilotMonitor.Functions.Services
             rule.IsBuiltIn = false;
             rule.IsCommunity = false;
             if (existing != null)
+            {
                 rule.CreatedAt = existing.CreatedAt;
+                // Author is stamped from the creator's token at create time and is
+                // immutable afterwards — updates (incl. cross-tenant GA edits) keep it.
+                rule.Author = existing.Author;
+            }
             rule.UpdatedAt = DateTime.UtcNow;
             return await _ruleRepo.StoreGatherRuleAsync(rule, tenantId);
         }

src/Backend/AutopilotMonitor.Functions/Services/Offboarding/TenantOffboardingHandler.cs

@@ -5,6 +5,7 @@ using System.Text.Json;
 using System.Threading;
 using System.Threading.Tasks;
 using AutopilotMonitor.Functions.DataAccess.TableStorage;
+using AutopilotMonitor.Functions.Functions.Admin;
 using AutopilotMonitor.Functions.Services.Deletion;
 using AutopilotMonitor.Shared;
 using AutopilotMonitor.Shared.DataAccess;
@@ -167,6 +168,7 @@ namespace AutopilotMonitor.Functions.Services.Offboarding
         private readonly ITenantOffboardingEnqueuer _reEnqueuer;
         private readonly OpsEventService _opsEvents;
         private readonly ITenantCustomsArchiveRepository _customsArchive;
+        private readonly IConfigRepository _configRepo;
         private readonly IOffboardFarewellEmailSender _farewellEmail;
         private readonly ILogger<TenantOffboardingHandler> _logger;
 
@@ -182,6 +184,7 @@ namespace AutopilotMonitor.Functions.Services.Offboarding
             ITenantOffboardingEnqueuer reEnqueuer,
             OpsEventService opsEvents,
             ITenantCustomsArchiveRepository customsArchive,
+            IConfigRepository configRepo,
             IOffboardFarewellEmailSender farewellEmail,
             ILogger<TenantOffboardingHandler> logger)
         {
@@ -196,6 +199,7 @@ namespace AutopilotMonitor.Functions.Services.Offboarding
             _reEnqueuer = reEnqueuer;
             _opsEvents = opsEvents;
             _customsArchive = customsArchive;
+            _configRepo = configRepo;
             _farewellEmail = farewellEmail;
             _logger = logger;
         }
@@ -582,6 +586,27 @@ namespace AutopilotMonitor.Functions.Services.Offboarding
         private async Task RunPostDrainPhasesAsync(
             OffboardingHistoryEntry history, string tenantId, CancellationToken ct)
         {
+            // 2.D-guard — un-offboard race check. The destructive phases below are only
+            // legal while the TenantConfiguration row is still the Phase-1 offboarding
+            // tombstone (or already deleted by a previous 2.F-final on this resume path;
+            // null passes). If an operator re-enabled the tenant mid-cascade — the config
+            // write paths refuse that, but this is the last line of defense — wiping now
+            // would destroy data of a live tenant, and sessions registered after the 2.B
+            // enumeration would survive as orphan rows while their Events/Signals
+            // partitions get wiped. Fail-closed; the Failed marker + OpsEvent route the
+            // conflict to an operator instead.
+            var currentConfig = await _configRepo.GetTenantConfigurationAsync(tenantId);
+            if (currentConfig != null && !TenantOffboardFunction.IsOffboardingTombstone(currentConfig))
+            {
+                var msg =
+                    $"TenantConfiguration is no longer the offboarding tombstone " +
+                    $"(Disabled={currentConfig.Disabled}, Reason={currentConfig.DisabledReason ?? "<null>"}) " +
+                    "— refusing to wipe a re-enabled tenant.";
+                await FailAsync(history, tenantId, "unoffboard_race", msg, ct);
+                throw new InvalidOperationException(
+                    $"TenantOffboarding fail-closed for tenant={tenantId} phase=unoffboard_race");
+            }
+
             // 2.D-archive (PR3.B §3) — Archive then wipe GatherRules / AnalyzeRules /
             // ImeLogPatterns. Fail-loud: a row that cannot be archived must NOT be wiped.
             // After the third iteration if the source still has rows, throw with

src/Backend/AutopilotMonitor.Functions/Services/TableStorageService.Sessions.cs

@@ -175,6 +175,15 @@ namespace AutopilotMonitor.Functions.Services
             if (!string.IsNullOrEmpty(reconcileReason))
                 indexEntity["ReconcileReason"] = reconcileReason;
 
+            // Soft-failure marker — rendered as the amber "completed with issues" badge in the
+            // index-served session list, so it must survive a StartedAt-shift full upsert too.
+            var espSoftFailure = sessionEntity.GetBoolean("EspSoftFailure");
+            if (espSoftFailure == true)
+                indexEntity["EspSoftFailure"] = true;
+            var completionSource = sessionEntity.GetString("CompletionSource");
+            if (!string.IsNullOrEmpty(completionSource))
+                indexEntity["CompletionSource"] = completionSource;
+
             // Device-validation path recorded at registration (AutopilotV1 / CorporateIdentifier /
             // DeviceAssociation / Bootstrap) — index-served session reads must carry it too.
             var validatedBy = sessionEntity.GetString("ValidatedBy");
@@ -647,6 +656,7 @@ namespace AutopilotMonitor.Functions.Services
                                 // fresh-read window isn't reverted by the Replace below (see guard).
                                 "Status", "CurrentPhase", "CompletedAt", "FailureReason", "ReconcileReason",
                                 "FailureSnapshotJson", "FailureSource", "AdminMarkedAction", "DurationSeconds",
+                                "EspSoftFailure", "CompletionSource",
                             });
                         freshEtag = freshResponse.Value.ETag;
                         freshEntity = freshResponse.Value;
@@ -686,7 +696,7 @@ namespace AutopilotMonitor.Functions.Services
                         || freshStatus == SessionStatus.Incomplete.ToString())
                     {
                         entity["Status"] = freshStatus;
-                        foreach (var col in new[] { "CurrentPhase", "CompletedAt", "FailureReason", "ReconcileReason", "FailureSnapshotJson", "FailureSource", "AdminMarkedAction", "DurationSeconds" })
+                        foreach (var col in new[] { "CurrentPhase", "CompletedAt", "FailureReason", "ReconcileReason", "FailureSnapshotJson", "FailureSource", "AdminMarkedAction", "DurationSeconds", "EspSoftFailure", "CompletionSource" })
                         {
                             if (freshEntity!.TryGetValue(col, out var val) && val is not null)
                                 entity[col] = val;
@@ -1590,7 +1600,7 @@ namespace AutopilotMonitor.Functions.Services
                 : null;
         }
 
-        public async Task<bool> UpdateSessionStatusAsync(string tenantId, string sessionId, SessionStatus status, EnrollmentPhase? currentPhase = null, string? failureReason = null, DateTime? completedAt = null, DateTime? earliestEventTimestamp = null, DateTime? latestEventTimestamp = null, bool? isPreProvisioned = null, bool? isUserDriven = null, DateTime? resumedAt = null, DateTime? stalledAt = null, bool clearStalledAt = false, bool clearFailureReason = false, string? failureSource = null, string? adminMarkedAction = null, string? failureSnapshotJson = null, bool allowTerminalReclassification = false)
+        public async Task<bool> UpdateSessionStatusAsync(string tenantId, string sessionId, SessionStatus status, EnrollmentPhase? currentPhase = null, string? failureReason = null, DateTime? completedAt = null, DateTime? earliestEventTimestamp = null, DateTime? latestEventTimestamp = null, bool? isPreProvisioned = null, bool? isUserDriven = null, DateTime? resumedAt = null, DateTime? stalledAt = null, bool clearStalledAt = false, bool clearFailureReason = false, string? failureSource = null, string? adminMarkedAction = null, string? failureSnapshotJson = null, bool allowTerminalReclassification = false, bool espSoftFailure = false, string? completionSource = null)
         {
             SecurityValidator.EnsureValidGuid(tenantId, nameof(tenantId));
             SecurityValidator.EnsureValidGuid(sessionId, nameof(sessionId));
@@ -1701,6 +1711,18 @@ namespace AutopilotMonitor.Functions.Services
                         var reconcileReason = ComputeReconcileReason(existingStatusStr, failureReason, adminMarkedAction);
                         if (reconcileReason != null)
                             update["ReconcileReason"] = reconcileReason;
+
+                        // Soft-failure marker (Continue-Anyway observation / advisory-defang):
+                        // the session succeeded, but the ESP demonstrably gave up on at least
+                        // one blocking item first. Rendered as the amber "completed with
+                        // issues" badge. Only ever stamped alongside a Succeeded transition —
+                        // absence means a clean success.
+                        if (espSoftFailure)
+                        {
+                            update["EspSoftFailure"] = true;
+                            if (!string.IsNullOrEmpty(completionSource))
+                                update["CompletionSource"] = completionSource;
+                        }
                     }
                     else if (status == SessionStatus.Failed)
                     {
@@ -1973,6 +1995,13 @@ namespace AutopilotMonitor.Functions.Services
                                 var forceReconcileReason = ComputeReconcileReason(freshStatusStr, failureReason, adminMarkedAction);
                                 if (forceReconcileReason != null)
                                     forceUpdate["ReconcileReason"] = forceReconcileReason;
+                                // Mirror the normal path's soft-failure marker.
+                                if (espSoftFailure)
+                                {
+                                    forceUpdate["EspSoftFailure"] = true;
+                                    if (!string.IsNullOrEmpty(completionSource))
+                                        forceUpdate["CompletionSource"] = completionSource;
+                                }
                             }
                             else if (status == SessionStatus.Failed)
                             {
@@ -2928,6 +2957,8 @@ namespace AutopilotMonitor.Functions.Services
                 FailureReason = entity.GetString("FailureReason") ?? string.Empty,
                 FailureSource = entity.GetString("FailureSource") ?? string.Empty,
                 ReconcileReason = entity.GetString("ReconcileReason") ?? string.Empty,
+                EspSoftFailure = entity.GetBoolean("EspSoftFailure") ?? false,
+                CompletionSource = entity.GetString("CompletionSource") ?? string.Empty,
                 AdminMarkedAction = entity.GetString("AdminMarkedAction"),
                 ValidatedBy = entity.GetString("ValidatedBy") ?? string.Empty,
                 PendingActionsJson = entity.GetString("PendingActionsJson") ?? string.Empty,

src/Backend/AutopilotMonitor.Functions/Services/TenantConfigPatchService.cs

@@ -4,6 +4,7 @@ using System.Linq;
 using System.Reflection;
 using System.Threading.Tasks;
 using AutopilotMonitor.Functions.DataAccess.TableStorage;
+using AutopilotMonitor.Functions.Functions.Admin;
 using AutopilotMonitor.Functions.Helpers;
 using AutopilotMonitor.Shared;
 using AutopilotMonitor.Shared.DataAccess;
@@ -131,6 +132,7 @@ namespace AutopilotMonitor.Functions.Services
             "Disabled", "DisabledReason", "DisabledUntil",
             "AllowInsecureAgentRequests", "BootstrapTokenEnabled",
             "UnrestrictedModeEnabled", "EntraAppRolesEnabled",
+            "EnableEspContinueAnywayObservation",
             "CustomRateLimitRequestsPerMinute", "CustomUserRateLimitRequestsPerMinute",
             "ValidateDeviceAssociation", "MaxNdjsonPayloadSizeMB",
         };
@@ -185,6 +187,14 @@ namespace AutopilotMonitor.Functions.Services
                     return PatchOutcome.Fail(PatchFailure.NotFound, $"Tenant {tenantId} has no configuration row.");
                 var (initial, etag) = read.Value;
 
+                // Mid-offboarding freeze: while the row is the offboarding tombstone, the
+                // cascade owns it. Any patch — a GA lifting Disabled included — would race
+                // the queued wipe: data written afterwards gets destroyed, and sessions
+                // registered after enumeration would survive as orphans.
+                if (TenantOffboardFunction.IsOffboardingTombstone(initial))
+                    return PatchOutcome.Fail(PatchFailure.ValidationFailed,
+                        "Tenant offboarding is in progress — the configuration is frozen until the cascade completes.");
+
                 // Patch onto a deep clone; MissingMemberHandling.Error is belt-and-braces on
                 // top of the key gate above. JSON null = explicit clear; omitted = untouched;
                 // camelCase input binds case-insensitively to the PascalCase model.
@@ -275,6 +285,12 @@ namespace AutopilotMonitor.Functions.Services
                     return PatchOutcome.Fail(PatchFailure.NotFound, $"Tenant {tenantId} has no configuration row.");
                 var (current, etag) = read.Value;
 
+                // Same mid-offboarding freeze as ApplyFieldPatchAsync: a revert restores a
+                // pre-offboarding snapshot with Disabled=false — the exact un-offboard race.
+                if (TenantOffboardFunction.IsOffboardingTombstone(current))
+                    return PatchOutcome.Fail(PatchFailure.ValidationFailed,
+                        "Tenant offboarding is in progress — the configuration is frozen until the cascade completes.");
+
                 var candidate = DeepClone(target);
                 // Identity is never restored from a snapshot; protected/system-owned fields
                 // only with the explicit GA opt-in.

src/McpServer/autopilot-monitor-mcp/src/__tests__/rule-validation.test.ts

@@ -137,6 +137,37 @@ describe('gather guardrails (agent matching semantics)', () => {
     expect(validateRuleDraft(spoof).valid).toBe(false);
   });
 
+  it('WMI: property projection of an allowed class is admitted', () => {
+    const single = validGather();
+    single.collectorType = 'wmi';
+    single.target = 'SELECT BatteryStatus FROM Win32_Battery';
+    expect(validateRuleDraft(single).valid).toBe(true);
+
+    const multi = validGather();
+    multi.collectorType = 'wmi';
+    multi.target = 'SELECT BatteryStatus, EstimatedChargeRemaining FROM Win32_Battery WHERE BatteryStatus = 1';
+    expect(validateRuleDraft(multi).valid).toBe(true);
+  });
+
+  it('WMI: projection of a disallowed class and star-in-list are rejected', () => {
+    const badClass = validGather();
+    badClass.collectorType = 'wmi';
+    badClass.target = 'SELECT Name FROM Win32_Process';
+    expect(validateRuleDraft(badClass).valid).toBe(false);
+
+    const starMix = validGather();
+    starMix.collectorType = 'wmi';
+    starMix.target = 'SELECT *, Name FROM Win32_BIOS';
+    expect(validateRuleDraft(starMix).valid).toBe(false);
+  });
+
+  it('WMI: leading whitespace is trimmed like the agent does (no false positive)', () => {
+    const padded = validGather();
+    padded.collectorType = 'wmi';
+    padded.target = '  SELECT * FROM Win32_BIOS';
+    expect(validateRuleDraft(padded).valid).toBe(true);
+  });
+
   it('commands match exactly — extra arguments are rejected', () => {
     const ok = validGather();
     ok.collectorType = 'command_allowlisted';

src/McpServer/autopilot-monitor-mcp/src/rule-validation.ts

@@ -67,6 +67,25 @@ const flatRegistryPrefixes = RULE_GUARDRAILS.registryPrefixes.flatMap((g) => [..
 const flatCommands = RULE_GUARDRAILS.allowedCommands.flatMap((g) => [...g.commands]);
 const flatEventLogChannels = RULE_GUARDRAILS.eventLogChannels.flatMap((g) => [...g.channels]);
 
+// Canonical WMI allowlist entry: exactly "SELECT * FROM <class>". Such entries
+// contribute their class name; any other entry (none today) stays on the legacy
+// literal-prefix path. Mirrors GatherRuleGuards.cs.
+const CANONICAL_WMI_ENTRY_REGEX = /^SELECT\s+\*\s+FROM\s+([A-Za-z_][A-Za-z0-9_]*)$/i;
+
+// WQL shape "SELECT <* | property-list> FROM <class>", class token bounded by
+// whitespace or end of string (trailing WHERE permitted, Win32_BIOSX cannot
+// pass as Win32_BIOS).
+const WMI_QUERY_REGEX =
+  /^SELECT\s+(?:\*|[A-Za-z_][A-Za-z0-9_]*(?:\s*,\s*[A-Za-z_][A-Za-z0-9_]*)*)\s+FROM\s+([A-Za-z_][A-Za-z0-9_]*)(?=\s|$)/i;
+
+const allowedWmiClasses = new Set<string>();
+const residualWmiPrefixes: string[] = [];
+for (const entry of RULE_GUARDRAILS.wmiQueryPrefixes) {
+  const m = CANONICAL_WMI_ENTRY_REGEX.exec(entry.trim());
+  if (m) allowedWmiClasses.add(m[1].toLowerCase());
+  else residualWmiPrefixes.push(entry);
+}
+
 /** Hard blocks enforced in agent CODE (not liftable via guardrails.json). Keep in
  * sync with GatherRuleGuards.cs — additions there are fail-safe (agent still blocks),
  * missing ones here just cost a pre-flight warning. */
@@ -121,15 +140,24 @@ function checkGatherTarget(collectorType: string, target: string): ValidationFin
       break;
     }
     case 'wmi': {
-      const t = target.toLowerCase();
-      const ok = RULE_GUARDRAILS.wmiQueryPrefixes.some((p) => {
-        const pl = p.toLowerCase();
-        // Whitespace-bounded: the query is the prefix or continues with whitespace
-        // (mirrors the agent's boundary check, prevents Win32_BIOSX spoofing).
-        return t === pl || (t.startsWith(pl) && /\s/.test(t.charAt(pl.length)));
-      });
+      // Trim like the agent does — an untrimmed target must not be a false positive.
+      const t = target.trim().toLowerCase();
+      const m = WMI_QUERY_REGEX.exec(t);
+      let ok = m !== null && allowedWmiClasses.has(m[1]);
       if (!ok) {
-        findings.push({ level: 'error', message: `guardrails: WMI query "${target}" does not start with any allowed query prefix (rule_guardrails → wmiQueryPrefixes).` });
+        // Non-canonical allowlist entries keep the legacy literal-prefix match.
+        ok = residualWmiPrefixes.some((p) => {
+          const pl = p.toLowerCase();
+          return t === pl || (t.startsWith(pl) && /\s/.test(t.charAt(pl.length)));
+        });
+      }
+      if (!ok) {
+        findings.push({
+          level: 'error',
+          message:
+            `guardrails: WMI query "${target}" is not allowed — use "SELECT <properties or *> FROM <class>" against a class from rule_guardrails → wmiQueryPrefixes. ` +
+            'Allowing additional classes requires a contribution to rules/guardrails.json in the product repository.',
+        });
       }
       break;
     }

src/Shared/AutopilotMonitor.DecisionCore/Engine/DeadlineNames.cs

@@ -111,6 +111,16 @@ namespace AutopilotMonitor.DecisionCore.Engine
         /// </summary>
         public const string EspAllowContinueAnyway = "espAllowContinueAnyway";
 
+        /// <summary>
+        /// On <c>EspConfigDetected</c>: "true" when the tenant opted into the
+        /// Continue-Anyway observation mode (operator-set
+        /// <c>EnableEspContinueAnywayObservation</c> tenant setting, stamped into the signal
+        /// payload by the agent from RemoteConfig — NOT an ESP registry fact). Only stamped
+        /// when enabled; missing → fact stays unset and a Device-phase ESP terminal failure
+        /// keeps the immediate hard-fail semantics.
+        /// </summary>
+        public const string EspContinueAnywayObservationEnabled = "espContinueAnywayObservationEnabled";
+
         /// <summary>On <c>HelloPolicyDetected</c>: "true" / "false". PR4 (882fef64 debrief).</summary>
         public const string HelloEnabled = "helloEnabled";
 

src/Shared/AutopilotMonitor.DecisionCore/Engine/DecisionEngine.Classic.cs

@@ -565,13 +565,23 @@ namespace AutopilotMonitor.DecisionCore.Engine
             // and surfaces hello_resolution via completion_waiting; the wait is bounded by an
             // already-armed HelloSafety, the tracker's Hello completion timer, or the
             // AdvisoryCompletion backstop.
+            // Continue-Anyway observation completion (2026-08-08): while a Device-phase
+            // observation advisory is live (see IsContinueAnywayObservationActive — ESP
+            // failed in DeviceSetup, tenant opted into observation, AccountSetup never
+            // entered), the strong post-AccountSetup gate is unsatisfiable by construction.
+            // The DAD-validated real-user desktop IS the observation's completion evidence:
+            // the user pressed "Continue anyway" and reached a working desktop. Complete
+            // eagerly here instead of waiting out the 60-min window so the session end
+            // reflects the real end.
+            var observationCompletion = IsContinueAnywayObservationActive(state);
+
             if (!helloAlreadyResolved
                 && HelloPolicyDisabledWithoutWizard(state.HelloPolicyEnabled, state.HelloWizardStartedUtc)
                 // desktopArrivedInFlight: the desktop fact was just recorded on the builder
                 // (line above) and is invisible to the pre-mutation state — arm C's "desktop
                 // arrives last" ordering (session a4537c36) must count it. Arms A/B ignore
                 // the flag entirely.
-                && ShouldTransitionToAwaitingHello(state, desktopArrivedInFlight: true))
+                && (ShouldTransitionToAwaitingHello(state, desktopArrivedInFlight: true) || observationCompletion))
             {
                 SynthesizeHelloSkipped(builder, signal);
 
@@ -600,12 +610,16 @@ namespace AutopilotMonitor.DecisionCore.Engine
 
                 // Completion gates (ARCH-F1): synthetic Hello + Desktop are recorded; complete
                 // through Finalizing when all gates are open, else defer until a gate releases.
+                // The observation trigger label keeps the two doors apart on the audit trail —
+                // the observation path deliberately bypassed the post-AccountSetup gate.
                 return CompleteThroughFinalizingOrDefer(
                     state: state,
                     signal: signal,
                     preparedBuilder: builder,
                     nextStepIndex: nextStep,
-                    trigger: nameof(DecisionSignalKind.DesktopArrived) + ":HelloDisabledFastPath",
+                    trigger: ShouldTransitionToAwaitingHello(state, desktopArrivedInFlight: true)
+                        ? nameof(DecisionSignalKind.DesktopArrived) + ":HelloDisabledFastPath"
+                        : nameof(DecisionSignalKind.DesktopArrived) + ":ContinueAnywayObservation",
                     leadingEffects: extraEffects);
             }
 
@@ -1029,16 +1043,37 @@ namespace AutopilotMonitor.DecisionCore.Engine
         /// </param>
         internal static DecisionEffect BuildEnrollmentCompleteEffect(DecisionState state, string completionTrigger)
         {
+            var trail = DecisionAuditTrailBuilder.Build(
+                postState: state,
+                decidedStage: state.Stage,
+                trigger: completionTrigger);
+
+            // Soft-failure honesty (2026-08-08): a session completing while an ESP-failure
+            // advisory is still unresolved reached the desktop, but the ESP demonstrably gave
+            // up on at least one blocking item first (Continue-Anyway observation, or the
+            // classic post-AccountSetup defang). Stamp that on the terminal event so the
+            // backend can mark the session "completed with issues" instead of presenting a
+            // clean success. A RESOLVED advisory (category re-ran to success after
+            // "Try again") is a genuine recovery and stays unstamped.
+            if (state.EspAdvisoryFailureRecordedUtc != null && state.EspAdvisoryFailureResolvedUtc == null)
+            {
+                trail["espSoftFailure"] = "true";
+                trail["completionSource"] = state.AccountSetupEnteredUtc == null
+                    ? "continue_anyway_observation"
+                    : "continue_anyway_post_accountsetup";
+                if (state.EspAdvisoryFailureCategory != null)
+                {
+                    trail["espSoftFailureCategory"] = state.EspAdvisoryFailureCategory.Value;
+                }
+            }
+
             return new DecisionEffect(
                 kind: DecisionEffectKind.EmitEventTimelineEntry,
                 parameters: new Dictionary<string, string>
                 {
                     ["eventType"] = SharedConstants.EventTypes.EnrollmentComplete,
                 },
-                typedPayload: DecisionAuditTrailBuilder.Build(
-                    postState: state,
-                    decidedStage: state.Stage,
-                    trigger: completionTrigger));
+                typedPayload: trail);
         }
     }
 }

src/Shared/AutopilotMonitor.DecisionCore/Engine/DecisionEngine.Edge.cs

@@ -237,20 +237,61 @@ namespace AutopilotMonitor.DecisionCore.Engine
             var observations = state.ScenarioObservations;
             var continueAnywayEnabled = observations?.EspAllowContinueAnyway?.Value == true;
             var advisoryEligible = continueAnywayEnabled && state.AccountSetupEnteredUtc != null;
+
+            // Continue-Anyway observation mode (tenant c9787ba2, session 53d1e9f6, 2026-08-08):
+            // a Device-phase ESP terminal failure (AccountSetup never entered) is a hard fail
+            // by default — the classic advisory gate above requires AccountSetupEnteredUtc.
+            // Fleets held at the ESP timeout wall by one slow blocking app then show 100%
+            // Failed although the profile allows "Continue anyway" and users demonstrably
+            // reach the desktop. When the tenant explicitly opted in
+            // (EnableEspContinueAnywayObservation → EspContinueAnywayObservationEnabled
+            // observation), defang this case too: arm a 60-min observation window instead of
+            // failing immediately. The DAD-validated real-user desktop + the Hello gate then
+            // detect the true end (see HandleDesktopArrivedV1's observation completion and
+            // the observation branch in HandleAdvisoryCompletionDeadlineFired); an expired
+            // window un-defangs to the original esp_terminal_failure. Scope: Classic /
+            // Unknown user-driven flows only — WhiteGlove / SelfDeploying / DevicePreparation
+            // have no Continue-Anyway user desktop to observe.
+            var observationEligible = !advisoryEligible
+                && continueAnywayEnabled
+                && observations?.EspContinueAnywayObservationEnabled?.Value == true
+                && state.AccountSetupEnteredUtc == null
+                && IsObservationEligibleScenario(state);
+
             var parameters = BuildEspFailureParameters(
                 signal: signal,
                 reason: reason,
-                advisoryPath: advisoryEligible,
+                advisoryPath: advisoryEligible || observationEligible,
                 observations: observations);
 
-            if (advisoryEligible)
+            if (advisoryEligible || observationEligible)
             {
-                return BuildAdvisoryStep(state, signal, nextStep, reason, parameters);
+                if (observationEligible)
+                {
+                    parameters["advisoryReason"] = "esp_failure_defanged_continueanyway_observation";
+                }
+                return BuildAdvisoryStep(state, signal, nextStep, reason, parameters, observationVariant: observationEligible);
             }
 
             return BuildFailedStep(state, signal, nextStep, reason, parameters);
         }
 
+        /// <summary>
+        /// Scenario gate for the Continue-Anyway observation defang: only Classic (or
+        /// still-unclassified) non-pre-provisioning flows qualify. WhiteGlove technician
+        /// sessions, SelfDeploying/kiosk profiles and DevicePreparation have no interactive
+        /// user who could press "Continue anyway" and produce a real-user desktop, so the
+        /// observation window could only ever expire — the immediate hard fail is more
+        /// truthful there.
+        /// </summary>
+        private static bool IsObservationEligibleScenario(DecisionState state)
+        {
+            var profile = state.ScenarioProfile;
+            return (profile.Mode == EnrollmentMode.Unknown || profile.Mode == EnrollmentMode.Classic)
+                && profile.PreProvisioningSide == PreProvisioningSide.None
+                && state.ScenarioObservations?.RegistrySelfDeployingProfile?.Value != true;
+        }
+
         /// <summary>
         /// Build the advisory-enrichment step that runs when an EspTerminalFailure duplicate
         /// arrives AFTER an advisory was already recorded AND the duplicate carries new
@@ -322,6 +363,17 @@ namespace AutopilotMonitor.DecisionCore.Engine
         /// </summary>
         private static readonly TimeSpan s_advisoryCompletionWindow = TimeSpan.FromMinutes(30);
 
+        /// <summary>
+        /// Observation window for the Continue-Anyway observation defang (Device-phase ESP
+        /// terminal failure, tenant opt-in — see <c>HandleEspTerminalFailureV1</c>). Longer
+        /// than <see cref="s_advisoryCompletionWindow"/> because the user must first notice
+        /// the failure screen, press "Continue anyway", sign in and reach the desktop —
+        /// or press "Try again" and let the ESP re-run the whole device phase. Fixed 60 min
+        /// (user decision 2026-08-08): well under the agent max lifetime, and every observed
+        /// session still gets a terminal verdict when the window expires.
+        /// </summary>
+        private static readonly TimeSpan s_observationCompletionWindow = TimeSpan.FromMinutes(60);
+
         /// <summary>
         /// Arming-time baseline keys carried on the AdvisoryCompletion deadline's
         /// <see cref="ActiveDeadline.FiresPayload"/> (session 1924092e enforcement-progress
@@ -343,10 +395,10 @@ namespace AutopilotMonitor.DecisionCore.Engine
         /// <see cref="HasEnforcementProgressSinceArming"/> can tell an idle dead-end from a
         /// still-progressing enrollment at fire time.
         /// </summary>
-        internal static ActiveDeadline BuildAdvisoryCompletionDeadline(DecisionState state, DecisionSignal signal) =>
+        internal static ActiveDeadline BuildAdvisoryCompletionDeadline(DecisionState state, DecisionSignal signal, TimeSpan? window = null) =>
             new ActiveDeadline(
                 name: DeadlineNames.AdvisoryCompletion,
-                dueAtUtc: EffectiveDeadlineBase(state, signal).Add(s_advisoryCompletionWindow),
+                dueAtUtc: EffectiveDeadlineBase(state, signal).Add(window ?? s_advisoryCompletionWindow),
                 firesSignalKind: DecisionSignalKind.DeadlineFired,
                 firesPayload: new Dictionary<string, string>
                 {
@@ -358,6 +410,23 @@ namespace AutopilotMonitor.DecisionCore.Engine
                             .ToString(CultureInfo.InvariantCulture),
                 });
 
+        /// <summary>
+        /// True while a Continue-Anyway observation advisory is live: an ESP-failure advisory
+        /// is recorded, not resolved, and AccountSetup was never entered. Only the
+        /// observation arming site in <see cref="HandleEspTerminalFailureV1"/> can produce
+        /// this shape — the classic advisory requires <see cref="DecisionState.AccountSetupEnteredUtc"/>
+        /// at arm time and neither fact is ever retracted — so no separate opt-in re-check is
+        /// needed. Consumed by the observation completion path in
+        /// <c>HandleDesktopArrivedV1</c> and the observation branches in
+        /// <see cref="HandleAdvisoryCompletionDeadlineFired"/>. Deliberately flips false when
+        /// AccountSetup IS entered later (user pressed "Try again" and the ESP re-ran into
+        /// the user phase) — from that point the classic advisory semantics apply.
+        /// </summary>
+        internal static bool IsContinueAnywayObservationActive(DecisionState state) =>
+            state.EspAdvisoryFailureRecordedUtc != null
+            && state.EspAdvisoryFailureResolvedUtc == null
+            && state.AccountSetupEnteredUtc == null;
+
         /// <summary>True when <see cref="DeadlineNames.AdvisoryCompletion"/> is armed in <paramref name="state"/>.</summary>
         internal static bool HasAdvisoryCompletionDeadline(DecisionState state)
             => FindAdvisoryCompletionDeadline(state) != null;
@@ -417,7 +486,8 @@ namespace AutopilotMonitor.DecisionCore.Engine
             DecisionSignal signal,
             int nextStep,
             string reason,
-            Dictionary<string, string> parameters)
+            Dictionary<string, string> parameters,
+            bool observationVariant = false)
         {
             // Session 4910a5a5: remember WHICH category failed so the recovery hook
             // (TryResolveAdvisoryOnCategoryRecovery) can later match a ProvisioningComplete
@@ -443,7 +513,13 @@ namespace AutopilotMonitor.DecisionCore.Engine
             // session either way (see HandleAdvisoryCompletionDeadlineFired). AddDeadline
             // replaces by name, so an earlier esp_exiting arm is intentionally re-based on the
             // (later) failure instant — the freshest dead-end signal owns the window.
-            var advisoryCompletion = BuildAdvisoryCompletionDeadline(state, signal);
+            //
+            // Observation variant (2026-08-08): the Device-phase Continue-Anyway observation
+            // uses the longer 60-min window — the user must traverse failure screen →
+            // Continue anyway → sign-in → desktop (or a full device-phase re-run via
+            // "Try again") before completion evidence can exist.
+            var advisoryCompletion = BuildAdvisoryCompletionDeadline(
+                state, signal, observationVariant ? s_observationCompletionWindow : (TimeSpan?)null);
             builder.AddDeadline(advisoryCompletion);
 
             var newState = builder.Build();
@@ -636,7 +712,17 @@ namespace AutopilotMonitor.DecisionCore.Engine
             // Shared with ShouldTransitionToAwaitingHello's proactive arm C (session a4537c36).
             var imeUserSessionGenuine = IsImeUserSessionGenuine(state);
 
-            if (desktopArrived && helloSatisfied && imeUserSessionGenuine)
+            // Continue-Anyway observation variant (2026-08-08): a Device-phase advisory —
+            // AccountSetup never entered — cannot satisfy the IME user-session gate by
+            // construction (the gate anchors on AccountSetupEnteredUtc). The DAD-validated
+            // real-user desktop plus the Hello gate ARE the completion evidence there: the
+            // user pressed "Continue anyway" and reached a working desktop. Only the
+            // observation arming site can produce this shape (the classic advisory requires
+            // AccountSetupEnteredUtc at arm time and the fact is never retracted), so the
+            // predicate needs no separate opt-in re-check.
+            var isObservationAdvisory = IsContinueAnywayObservationActive(state);
+
+            if (desktopArrived && helloSatisfied && (imeUserSessionGenuine || isObservationAdvisory))
             {
                 if (state.HelloResolvedUtc == null)
                 {
@@ -669,7 +755,7 @@ namespace AutopilotMonitor.DecisionCore.Engine
             // the Hello wizard right now. Promote to AwaitingHello and let a real
             // HelloResolved — or the HelloSafety timeout — decide instead.
             if (desktopArrived
-                && imeUserSessionGenuine
+                && (imeUserSessionGenuine || isObservationAdvisory)
                 && state.HelloResolvedUtc == null
                 && state.HelloPolicyEnabled?.Value == false
                 && state.HelloWizardStartedUtc != null)
@@ -722,7 +808,13 @@ namespace AutopilotMonitor.DecisionCore.Engine
             // — the failure DID un-happen, so a live AccountSetup must not be failed on the
             // stale advisory). Convergent either way: each re-arm renews the baselines, so a
             // session that stalls after the recovery still fails one window later.
-            if ((!hasAdvisoryAnchor || advisoryResolved) && HasEnforcementProgressSinceArming(state, armedDeadline!))
+            // The observation variant participates too (2026-08-08): its whole point is to
+            // keep watching while there is evidence of life — a user who pressed "Try again"
+            // late in the window keeps apps reaching terminal states, and failing mid-retry
+            // would be exactly the false positive the guard exists to prevent. Convergent:
+            // each re-arm (30 min) demands NEW progress.
+            if ((!hasAdvisoryAnchor || advisoryResolved || isObservationAdvisory)
+                && HasEnforcementProgressSinceArming(state, armedDeadline!))
             {
                 var rearmedDeadline = BuildAdvisoryCompletionDeadline(state, signal);
                 builder.AddDeadline(rearmedDeadline);
@@ -778,7 +870,7 @@ namespace AutopilotMonitor.DecisionCore.Engine
             // through Finalizing. Deliberately does NOT require the IME user-session gate —
             // the FP sessions never saw an IME user-session completion (that gate guards
             // completion evidence, not user presence; presence is the DAD-validated desktop).
-            if ((!hasAdvisoryAnchor || advisoryResolved)
+            if ((!hasAdvisoryAnchor || advisoryResolved || isObservationAdvisory)
                 && desktopArrived
                 && state.HelloResolvedUtc == null
                 && state.HelloPolicyEnabled == null
@@ -861,7 +953,14 @@ namespace AutopilotMonitor.DecisionCore.Engine
                     reason: reason,
                     advisoryPath: false,
                     observations: state.ScenarioObservations);
-                parameters["advisoryReason"] = "advisory_completion_window_expired_without_completion_evidence";
+                // Observation variant: the 60-min Continue-Anyway observation produced no
+                // real-user desktop — either nobody pressed the button or the device truly
+                // is stuck at the failure screen. The un-defang is identical (the failure
+                // cause IS the ESP failure), only the advisoryReason tells the two windows
+                // apart on the timeline.
+                parameters["advisoryReason"] = isObservationAdvisory
+                    ? "continue_anyway_observation_window_expired_without_completion_evidence"
+                    : "advisory_completion_window_expired_without_completion_evidence";
                 builder.WithLastFailureTrigger(nameof(DecisionSignalKind.EspTerminalFailure), signal.SessionSignalOrdinal);
             }
             else

src/Shared/AutopilotMonitor.DecisionCore/State/EnrollmentScenarioObservations.cs

@@ -68,6 +68,18 @@ namespace AutopilotMonitor.DecisionCore.State
         /// </summary>
         public SignalFact<bool>? EspAllowContinueAnyway { get; init; }
 
+        /// <summary>
+        /// Tenant-config opt-in from <see cref="Signals.DecisionSignalKind.EspConfigDetected"/>
+        /// (payload <c>espContinueAnywayObservationEnabled</c>, stamped by the agent from
+        /// <c>RemoteConfig.EnableEspContinueAnywayObservation</c> — an operator-set tenant
+        /// setting, not an ESP registry fact). When <c>true</c> AND the ESP profile allows
+        /// "Continue anyway", a Device-phase ESP terminal failure (AccountSetup never entered)
+        /// is defanged into an observation advisory instead of failing the session immediately
+        /// — see <c>HandleEspTerminalFailureV1</c>. Default (absent/false) keeps the immediate
+        /// hard-fail semantics. Set-once.
+        /// </summary>
+        public SignalFact<bool>? EspContinueAnywayObservationEnabled { get; init; }
+
         /// <summary>
         /// Raw registry fact from <see cref="Signals.DecisionSignalKind.EnrollmentFactsObserved"/>
         /// (payload <c>isSelfDeployingProfile</c>, from <c>CloudAssignedOobeConfig</c> bits
@@ -126,6 +138,11 @@ namespace AutopilotMonitor.DecisionCore.State
                 ? this
                 : this with { EspAllowContinueAnyway = new SignalFact<bool>(value, sourceSignalOrdinal) };
 
+        public EnrollmentScenarioObservations WithEspContinueAnywayObservationEnabled(bool value, long sourceSignalOrdinal) =>
+            EspContinueAnywayObservationEnabled != null
+                ? this
+                : this with { EspContinueAnywayObservationEnabled = new SignalFact<bool>(value, sourceSignalOrdinal) };
+
         public EnrollmentScenarioObservations WithRegistrySelfDeployingProfile(bool value, long sourceSignalOrdinal) =>
             RegistrySelfDeployingProfile != null
                 ? this

src/Shared/AutopilotMonitor.DecisionCore/State/EnrollmentScenarioProfileUpdater.cs

@@ -226,6 +226,11 @@ namespace AutopilotMonitor.DecisionCore.State
                 {
                     newObservations = newObservations.WithEspAllowContinueAnyway(allowContinue, signal.SessionSignalOrdinal);
                 }
+                if (signal.Payload.TryGetValue(SignalPayloadKeys.EspContinueAnywayObservationEnabled, out var rawObservation)
+                    && bool.TryParse(rawObservation, out var observationEnabled))
+                {
+                    newObservations = newObservations.WithEspContinueAnywayObservationEnabled(observationEnabled, signal.SessionSignalOrdinal);
+                }
             }
 
             var newProfile = currentProfile;

src/Shared/AutopilotMonitor.Shared/DataAccess/ISessionRepository.cs

@@ -81,7 +81,8 @@ namespace AutopilotMonitor.Shared.DataAccess
             bool? isUserDriven = null, DateTime? resumedAt = null,
             DateTime? stalledAt = null, bool clearStalledAt = false, bool clearFailureReason = false,
             string? failureSource = null, string? adminMarkedAction = null,
-            string? failureSnapshotJson = null, bool allowTerminalReclassification = false);
+            string? failureSnapshotJson = null, bool allowTerminalReclassification = false,
+            bool espSoftFailure = false, string? completionSource = null);
         /// <summary>
         /// Increments per-session counters via read-modify-write. Returns the post-merge session
         /// snapshot (the RMW read with the applied increments) so hot-path callers can skip a

src/Shared/AutopilotMonitor.Shared/Models/Config/AgentConfigResponse.cs

@@ -46,6 +46,14 @@ namespace AutopilotMonitor.Shared.Models
         /// </summary>
         public bool EnableGatherRuleDebugLog { get; set; } = false;
 
+        /// <summary>
+        /// Continue-Anyway observation mode (operator-set tenant opt-in): a Device-phase ESP
+        /// terminal failure on a Continue-Anyway-enabled profile is observed for up to 60
+        /// minutes instead of failing immediately — the real-user desktop then completes the
+        /// session with an esp-soft-failure marker. Default false = immediate hard fail.
+        /// </summary>
+        public bool EnableEspContinueAnywayObservation { get; set; } = false;
+
         public CollectorConfiguration Collectors { get; set; } = default!;
 
         /// <summary>

src/Shared/AutopilotMonitor.Shared/Models/Config/TenantConfiguration.cs

@@ -382,6 +382,17 @@ namespace AutopilotMonitor.Shared.Models
         /// </summary>
         public bool? EnableGatherRuleDebugLog { get; set; }
 
+        /// <summary>
+        /// Continue-Anyway observation mode: when true AND the ESP profile allows
+        /// "Continue anyway", a Device-phase ESP terminal failure (AccountSetup never
+        /// entered) does not fail the session immediately — the agent keeps monitoring for
+        /// up to 60 minutes and completes with an esp-soft-failure marker once the
+        /// DAD-validated real-user desktop (plus the Hello gate) proves the user continued;
+        /// an expired window fails with the original esp_terminal_failure. Operator-set only
+        /// (not exposed in the tenant admin UI). null = use agent default (false).
+        /// </summary>
+        public bool? EnableEspContinueAnywayObservation { get; set; }
+
         /// <summary>
         /// Log verbosity level override for this tenant's agents.
         /// null = use agent default ("Info"). Values: "Info", "Debug", "Verbose", "Trace".

src/Shared/AutopilotMonitor.Shared/Models/GeographicApiModels.cs

@@ -131,6 +131,8 @@ namespace AutopilotMonitor.Shared.Models
                 Status = s.Status,
                 FailureReason = s.FailureReason,
                 FailureSource = s.FailureSource,
+                EspSoftFailure = s.EspSoftFailure,
+                CompletionSource = s.CompletionSource,
                 EventCount = s.EventCount,
                 DurationSeconds = s.DurationSeconds,
                 EnrollmentType = s.EnrollmentType,

src/Shared/AutopilotMonitor.Shared/Models/SessionApiModels.cs

@@ -231,6 +231,26 @@ namespace AutopilotMonitor.Shared.Models
         /// </summary>
         public string ReconcileReason { get; set; } = string.Empty;
 
+        /// <summary>
+        /// True only on a Succeeded session that completed while an ESP-failure advisory was
+        /// still unresolved: the ESP demonstrably gave up on at least one blocking item
+        /// (typically the 30-min timeout wall) and the user most likely pressed
+        /// "Continue anyway" — the device reached a working desktop, but not everything the
+        /// ESP tracked finished inside its window. Rendered as the amber "completed with
+        /// issues" badge. False/absent = clean success.
+        /// </summary>
+        public bool EspSoftFailure { get; set; }
+
+        /// <summary>
+        /// How a soft-failure completion was detected. Values:
+        ///   - "continue_anyway_observation": Device-phase ESP failure, tenant opted into the
+        ///     Continue-Anyway observation window, real-user desktop arrived inside it
+        ///   - "continue_anyway_post_accountsetup": classic advisory defang (AccountSetup had
+        ///     already been entered when the ESP failure fired)
+        /// Empty on clean successes.
+        /// </summary>
+        public string CompletionSource { get; set; } = string.Empty;
+
         /// <summary>
         /// Non-null only when an administrator explicitly flipped the session via the portal
         /// (MarkSessionSucceeded / MarkSessionFailed). Values: <c>null</c> (default, agent-driven),

src/Web/autopilot-monitor-web/app/admin/components/TenantManagementSection.tsx

@@ -33,6 +33,8 @@ export interface TenantConfiguration {
   bootstrapTokenEnabled?: boolean;
   unrestrictedModeEnabled?: boolean;
   entraAppRolesEnabled?: boolean;
+  /** Operator-set (GA-only): Device-phase ESP failures on Continue-Anyway profiles get a 60-min observation instead of an immediate hard fail. */
+  enableEspContinueAnywayObservation?: boolean;
   dataRetentionDays: number;
   sessionTimeoutHours: number;
   planTier?: string;
@@ -1005,6 +1007,24 @@ function TenantManagementSectionInner({
                 <span className="text-sm font-medium text-gray-700">Enable Unrestricted Mode</span>
               </label>
 
+              <div>
+                <label className="flex items-center space-x-2 cursor-pointer">
+                  <input
+                    type="checkbox"
+                    checked={editingTenant.enableEspContinueAnywayObservation ?? false}
+                    onChange={(e) => setEditingTenant({ ...editingTenant, enableEspContinueAnywayObservation: e.target.checked })}
+                    className="w-4 h-4 text-teal-600 border-gray-300 rounded focus:ring-teal-500"
+                  />
+                  <span className="text-sm font-medium text-gray-700">Enable Continue-Anyway Observation</span>
+                </label>
+                <p className="text-xs text-gray-400 mt-1 ml-6">
+                  Device-phase ESP terminal failures on profiles that allow &quot;Continue anyway&quot; are
+                  observed for up to 60 min instead of failing immediately: a real-user desktop completes
+                  the session as Succeeded with an amber &quot;with issues&quot; badge; otherwise it fails with
+                  the original reason. Needs agent with ConfigVersion 37+; applies to new sessions only.
+                </p>
+              </div>
+
               <div>
                 <label className="flex items-center space-x-2 cursor-pointer">
                   <input

src/Web/autopilot-monitor-web/app/admin/tenants/sections/SectionTenantConfigReport.tsx

@@ -59,6 +59,7 @@ interface TenantConfig {
   ntpServer?: string | null;
   enableImeMatchLog?: boolean | null;
   enableGatherRuleDebugLog?: boolean | null;
+  enableEspContinueAnywayObservation?: boolean | null;
   logLevel?: string | null;
   maxBatchSize?: number | null;
   showEnrollmentSummary?: boolean | null;
@@ -132,6 +133,7 @@ const DEFAULTS: Record<string, unknown> = {
   ntpServer: 'time.windows.com',
   enableImeMatchLog: null,
   enableGatherRuleDebugLog: null,
+  enableEspContinueAnywayObservation: null,
   logLevel: null,
   maxBatchSize: null,
   showEnrollmentSummary: null,
@@ -171,6 +173,7 @@ const RUNTIME_DEFAULTS: Record<string, unknown> = {
   ntpServer: 'time.windows.com',
   enableImeMatchLog: false,
   enableGatherRuleDebugLog: false,
+  enableEspContinueAnywayObservation: false,
   maxAuthFailures: 5,
   authFailureTimeoutMinutes: 0,
   logLevel: 'Info',
@@ -300,6 +303,7 @@ function computeRuntime(c: TenantConfig): Record<string, unknown> {
     ntpServer: c.ntpServer ?? 'time.windows.com',
     enableImeMatchLog: c.enableImeMatchLog ?? false,
     enableGatherRuleDebugLog: c.enableGatherRuleDebugLog ?? false,
+    enableEspContinueAnywayObservation: c.enableEspContinueAnywayObservation ?? false,
     maxAuthFailures: c.maxAuthFailures ?? 5,
     authFailureTimeoutMinutes: c.authFailureTimeoutMinutes ?? 0,
     logLevel: c.logLevel ?? 'Info',
@@ -581,6 +585,7 @@ export function SectionTenantConfigReport() {
                 <ConfigRow label="NTP Server" value={config.ntpServer} configKey="ntpServer" defaults={DEFAULTS} />
                 <ConfigRow label="IME Match Log" value={config.enableImeMatchLog} configKey="enableImeMatchLog" defaults={DEFAULTS} />
                 <ConfigRow label="Gather Rule Debug Log" value={config.enableGatherRuleDebugLog} configKey="enableGatherRuleDebugLog" defaults={DEFAULTS} />
+                <ConfigRow label="Continue-Anyway Observation" value={config.enableEspContinueAnywayObservation} configKey="enableEspContinueAnywayObservation" defaults={DEFAULTS} />
                 <ConfigRow label="Log Level" value={config.logLevel} configKey="logLevel" defaults={DEFAULTS} />
                 <ConfigRow label="Max Batch Size" value={config.maxBatchSize} configKey="maxBatchSize" defaults={DEFAULTS} />
                 <ConfigRow label="Show Script Output" value={config.showScriptOutput} configKey="showScriptOutput" defaults={DEFAULTS} />
@@ -654,6 +659,7 @@ export function SectionTenantConfigReport() {
                     <ConfigRow label="NTP Server" value={runtime.ntpServer} configKey="ntpServer" defaults={RUNTIME_DEFAULTS} />
                     <ConfigRow label="IME Match Log" value={runtime.enableImeMatchLog} configKey="enableImeMatchLog" defaults={RUNTIME_DEFAULTS} />
                     <ConfigRow label="Gather Rule Debug Log" value={runtime.enableGatherRuleDebugLog} configKey="enableGatherRuleDebugLog" defaults={RUNTIME_DEFAULTS} />
+                    <ConfigRow label="Continue-Anyway Observation" value={runtime.enableEspContinueAnywayObservation} configKey="enableEspContinueAnywayObservation" defaults={RUNTIME_DEFAULTS} />
                     <ConfigRow label="Log Level" value={runtime.logLevel} configKey="logLevel" defaults={RUNTIME_DEFAULTS} />
                     <ConfigRow label="Send Trace Events" value={runtime.sendTraceEvents} configKey="sendTraceEvents" defaults={RUNTIME_DEFAULTS} />
                     <ConfigRow label="Unrestricted Mode" value={runtime.unrestrictedMode} configKey="unrestrictedMode" defaults={RUNTIME_DEFAULTS} />

src/Web/autopilot-monitor-web/app/dashboard/components/SessionTable.tsx

@@ -862,7 +862,7 @@ function SessionCell({
           {/* Qualifier pills stack BELOW the status badge (flex-col) so extra badges
               (Hybrid / Self-Deploying / Blocked) never widen the status column. */}
           <div className="flex flex-col items-start gap-1">
-            <SessionStatusBadge status={session.status} failureReason={session.failureReason} adminMarkedAction={session.adminMarkedAction} reconcileReason={session.reconcileReason} />
+            <SessionStatusBadge status={session.status} failureReason={session.failureReason} adminMarkedAction={session.adminMarkedAction} reconcileReason={session.reconcileReason} espSoftFailure={session.espSoftFailure} />
             {session.isHybridJoin && (
               <span
                 className="px-2 inline-flex items-center gap-1 text-xs leading-5 font-semibold rounded-full bg-purple-100 text-purple-800"

src/Web/autopilot-monitor-web/app/gather-rules/__tests__/targetBlocked.test.ts

@@ -0,0 +1,35 @@
+import { describe, it, expect } from "vitest";
+import { targetBlocked } from "../types";
+
+// The enable gate for gather rules: a CUSTOM rule whose target fails guardrail
+// validation must not be enableable — the agent would block it on every device.
+
+const custom = { isBuiltIn: false, isCommunity: false };
+
+describe("targetBlocked", () => {
+  it("blocks a custom rule with a disallowed WMI query", () => {
+    expect(targetBlocked({ ...custom, collectorType: "wmi", target: "SELECT * FROM Win32_Process" }, false)).toBe(true);
+  });
+
+  it("does not block an allowed projection query", () => {
+    expect(targetBlocked({ ...custom, collectorType: "wmi", target: "SELECT BatteryStatus FROM Win32_Battery" }, false)).toBe(false);
+  });
+
+  it("never blocks built-in or community rules", () => {
+    expect(targetBlocked({ isBuiltIn: true, isCommunity: false, collectorType: "wmi", target: "SELECT * FROM Win32_Process" }, false)).toBe(false);
+    expect(targetBlocked({ isBuiltIn: false, isCommunity: true, collectorType: "wmi", target: "SELECT * FROM Win32_Process" }, false)).toBe(false);
+  });
+
+  it("respects unrestricted mode", () => {
+    expect(targetBlocked({ ...custom, collectorType: "wmi", target: "SELECT * FROM Win32_Process" }, true)).toBe(false);
+  });
+
+  it("does not block an empty target (required-field validation owns that)", () => {
+    expect(targetBlocked({ ...custom, collectorType: "wmi", target: "" }, false)).toBe(false);
+  });
+
+  it("blocks disallowed targets of other collector types too", () => {
+    expect(targetBlocked({ ...custom, collectorType: "file", target: "C:\\Users\\alice\\secret.txt" }, false)).toBe(true);
+    expect(targetBlocked({ ...custom, collectorType: "eventlog", target: "Security" }, false)).toBe(true);
+  });
+});

src/Web/autopilot-monitor-web/app/gather-rules/components/GatherRuleCard.tsx

@@ -68,6 +68,11 @@ export function GatherRuleCard({
     [rule.collectorType, rule.target, unrestrictedMode]
   );
 
+  // Custom rule whose target the agent would block on every device (security_warning,
+  // no data): the portal refuses to enable it; disabling always stays possible.
+  const isBlockedTarget = canEdit && targetValidation !== null && !targetValidation.allowed;
+  const enableBlocked = !rule.enabled && isBlockedTarget;
+
   return (
     <div
       className={`bg-white rounded-lg shadow border transition-all ${
@@ -99,13 +104,15 @@ export function GatherRuleCard({
                 e.stopPropagation();
                 onToggle();
               }}
-              disabled={togglingRule === rule.ruleId}
+              disabled={togglingRule === rule.ruleId || enableBlocked}
               className={`relative inline-flex h-6 w-11 flex-shrink-0 items-center rounded-full transition-colors ${
-                togglingRule === rule.ruleId
+                togglingRule === rule.ruleId || enableBlocked
                   ? "opacity-50 cursor-not-allowed"
                   : "cursor-pointer"
               } ${rule.enabled ? "bg-emerald-500" : "bg-gray-300"}`}
-              title={rule.enabled ? "Disable rule" : "Enable rule"}
+              title={enableBlocked
+                ? "Target is not allowed by the collection guardrails — fix the target before enabling"
+                : rule.enabled ? "Disable rule" : "Enable rule"}
             >
               <span
                 className={`inline-block h-4 w-4 transform rounded-full bg-white transition-transform ${
@@ -150,6 +157,16 @@ export function GatherRuleCard({
             {rule.isBuiltIn ? "Built-in" : rule.isCommunity ? "Community" : "Custom"}
           </span>
 
+          {/* Enabled rule the agent blocks on every device — visible without expanding */}
+          {rule.enabled && isBlockedTarget && (
+            <span
+              className="inline-flex items-center px-2 py-0.5 rounded text-xs font-medium bg-red-50 text-red-600 border border-red-200 flex-shrink-0"
+              title={targetValidation?.reason}
+            >
+              Blocked on devices
+            </span>
+          )}
+
           {/* Expand/Collapse Arrow */}
           <svg
             className={`w-5 h-5 text-gray-400 transition-transform flex-shrink-0 ml-auto sm:ml-0 ${
@@ -263,6 +280,20 @@ export function GatherRuleCard({
                   {rule.target}
                 </code>
                 <ValidationIndicator result={targetValidation} className="ml-2" />
+                {isBlockedTarget && (
+                  <p className="mt-1.5 text-xs text-gray-500">
+                    This target isn&apos;t on the collection allowlist. Think it should be?{" "}
+                    <a
+                      href="https://github.com/okieselbach/AutopilotMonitor/issues"
+                      target="_blank"
+                      rel="noopener noreferrer"
+                      className="text-indigo-600 hover:text-indigo-800 underline"
+                    >
+                      Request an addition via a GitHub issue
+                    </a>
+                    .
+                  </p>
+                )}
               </div>
               {rule.parameters && Object.keys(rule.parameters).length > 0 && (
                 <div className="text-sm">

src/Web/autopilot-monitor-web/app/gather-rules/components/GatherRuleFormFields.tsx

@@ -145,6 +145,21 @@ export function GatherRuleFormFields({ form, setForm, showRuleId, unrestrictedMo
           <p className="text-xs text-gray-400">{TARGET_HINTS[form.collectorType] || "Registry path, WMI class, event log name, file path, or command depending on collector type"}</p>
           <ValidationIndicator result={targetValidation} />
         </div>
+        {targetValidation && !targetValidation.allowed && (
+          <p className="mt-1 text-xs text-gray-500">
+            This target isn&apos;t on the collection allowlist — the rule can be saved but not enabled.
+            Think it should be allowed?{" "}
+            <a
+              href="https://github.com/okieselbach/AutopilotMonitor/issues"
+              target="_blank"
+              rel="noopener noreferrer"
+              className="text-indigo-600 hover:text-indigo-800 underline"
+            >
+              Request an addition via a GitHub issue
+            </a>
+            .
+          </p>
+        )}
       </div>
 
       {/* Registry: optional Value Name */}

src/Web/autopilot-monitor-web/app/gather-rules/page.tsx

@@ -14,7 +14,7 @@ import { FormJsonToggle, JsonModeToggleButtons } from "@/components/rules/FormJs
 import { useAuthenticatedFetch, useNotificationMessages, useGlobalAdminScope } from "@/hooks";
 import { GlobalAdminBanner, globalAdminSubtitle } from "@/components/GlobalAdminBanner";
 import { TenantScopeSelector } from "@/components/TenantScopeSelector";
-import { GatherRule, NewRuleForm, EMPTY_FORM, PHASE_TRIGGERS, buildScopeFields, validateScopeSelection, withDerivedScopeMode } from "./types";
+import { GatherRule, NewRuleForm, EMPTY_FORM, PHASE_TRIGGERS, buildScopeFields, targetBlocked, validateScopeSelection, withDerivedScopeMode } from "./types";
 import { GatherRuleFormFields } from "./components/GatherRuleFormFields";
 import { GatherRuleCard } from "./components/GatherRuleCard";
 import { DOCS_URL } from "@/utils/config";
@@ -125,6 +125,12 @@ export default function GatherRulesPage() {
   }, [effectiveTenantId, getAccessToken]);
 
   const handleToggleRule = async (rule: GatherRule) => {
+    // Defense in depth next to the disabled toggle button: never enable a rule the
+    // agent would block on every device anyway. Disabling always stays possible.
+    if (!rule.enabled && targetBlocked(rule, unrestrictedMode)) {
+      showError(`Rule "${rule.title}" cannot be enabled — its target is not allowed by the collection guardrails.`);
+      return;
+    }
     setTogglingRule(rule.ruleId);
     const result = await mutate(
       gatherRuleWriteUrl(rule.ruleId),
@@ -214,6 +220,9 @@ export default function GatherRulesPage() {
     }
 
     setCreating(true);
+    // Without an explicit enabled the backend model defaults to true — a rule the agent
+    // would block on every device must land disabled instead.
+    const blocked = targetBlocked(form, unrestrictedMode);
     const payload = {
       ruleId: form.ruleId,
       title: form.title,
@@ -229,6 +238,7 @@ export default function GatherRulesPage() {
       ...buildScopeFields(form),
       outputEventType: form.outputEventType,
       outputSeverity: form.outputSeverity,
+      enabled: !blocked,
     };
 
     const result = await mutate(
@@ -241,7 +251,9 @@ export default function GatherRulesPage() {
     );
     if (result !== null) {
       trackEvent("rule_created", { ruleType: "gather" });
-      showSuccess(`Rule "${form.title}" created`);
+      showSuccess(blocked
+        ? `Rule "${form.title}" created (disabled — its target is not allowed by the collection guardrails)`
+        : `Rule "${form.title}" created`);
       setShowCreateForm(false);
       setNewRule({ ...EMPTY_FORM });
       setJsonModeCreate(false);
@@ -306,6 +318,9 @@ export default function GatherRulesPage() {
     }
 
     setSaving(true);
+    // An edit that makes the target invalid force-disables the rule — the agent would
+    // block it on every device, so leaving it enabled only produces security warnings.
+    const forcedOff = rule.enabled && targetBlocked(form, unrestrictedMode);
     const payload = {
       title: form.title,
       description: form.description,
@@ -321,8 +336,7 @@ export default function GatherRulesPage() {
       outputEventType: form.outputEventType,
       outputSeverity: form.outputSeverity,
       version: bumpVersion(rule.version),
-      enabled: rule.enabled,
-      author: rule.author,
+      enabled: rule.enabled && !forcedOff,
       createdAt: rule.createdAt,
       // Same wipe class as the toggle bug: the full-replace PUT must carry tags or they vanish.
       tags: rule.tags ?? [],
@@ -340,7 +354,9 @@ export default function GatherRulesPage() {
     );
     if (result !== null) {
       trackEvent("rule_modified", { ruleType: "gather" });
-      showSuccess(`Rule "${form.title}" saved`);
+      showSuccess(forcedOff
+        ? `Rule "${form.title}" saved and disabled — its target is not allowed by the collection guardrails`
+        : `Rule "${form.title}" saved`);
       setEditingRuleId(null);
       setJsonModeEdit(false);
       setJsonError(null);

src/Web/autopilot-monitor-web/app/gather-rules/types.ts

@@ -1,3 +1,5 @@
+import { validateGatherRuleTarget } from "@/utils/guardValidation";
+
 export interface GatherRule {
   ruleId: string;
   title: string;
@@ -126,7 +128,7 @@ export const TARGET_PLACEHOLDERS: Record<string, string> = {
 export const TARGET_HINTS: Record<string, string> = {
   registry: "Full registry path including hive (HKLM, HKCU). The agent reads values from this key.",
   eventlog: "Event log name — supports operational/analytic logs like Microsoft-Windows-Shell-Core/Operational.",
-  wmi: "Full WQL query (SELECT * FROM ...). Must use an allowed WMI class.",
+  wmi: "Full WQL query — SELECT * or a comma-separated property list from an allowed WMI class, e.g. SELECT BatteryStatus FROM Win32_Battery. Pairs well with Emit Mode \"On change\": polling a single property only emits when that property changes.",
   file: "File path. Environment variables like %ProgramData% are supported. Must be within allowed directories.",
   command_allowlisted: "Exact command string from the agent's allowlist. Custom commands are not permitted.",
   logparser: "Path to a log file. Supports wildcards (* and ?) in the filename, e.g. AppWorkload-*.log. Environment variables are expanded.",
@@ -203,6 +205,23 @@ export function supportsEmitMode(form: Pick<NewRuleForm, "trigger" | "triggerPha
   return !firesExactlyOnce(form);
 }
 
+/**
+ * True when a CUSTOM rule's target fails guardrail validation — the agent would block
+ * it on every device (security_warning, no data), so the portal refuses to enable it.
+ * Built-in/community rules are catalog-validated and never gated; unrestrictedMode is
+ * respected because the validator returns allowed=true for it.
+ */
+export function targetBlocked(
+  rule: Pick<GatherRule, "collectorType" | "target"> & Partial<Pick<GatherRule, "isBuiltIn" | "isCommunity">>,
+  unrestrictedMode: boolean
+): boolean {
+  if (rule.isBuiltIn || rule.isCommunity) return false;
+  if (!rule.target?.trim()) return false;
+  // null = collector type the validator doesn't know — no verdict, don't gate.
+  const result = validateGatherRuleTarget(rule.collectorType, rule.target, unrestrictedMode);
+  return result !== null && !result.allowed;
+}
+
 /**
  * Rejects a scope mode that was selected but left unfilled. Without this the payload
  * silently sends null (= unrestricted) and the rule runs everywhere while the form still

src/Web/autopilot-monitor-web/app/geographic-performance/sessions/page.tsx

@@ -31,6 +31,8 @@ interface SessionSummary {
   completedAt: string | null;
   status: string;
   failureReason: string | null;
+  /** Succeeded despite an unresolved ESP failure (Continue-Anyway) — amber "with issues" badge. */
+  espSoftFailure?: boolean;
   durationSeconds: number | null;
   enrollmentType: string;
   /** Session-wide average agent→backend HTTP round-trip (ms); absent for pre-feature agents. */
@@ -360,7 +362,7 @@ function LocationSessionsContent() {
                           {session.model || "—"}
                         </td>
                         <td className="px-4 py-3 text-sm">
-                          <SessionStatusBadge status={session.status} failureReason={session.failureReason} />
+                          <SessionStatusBadge status={session.status} failureReason={session.failureReason} espSoftFailure={session.espSoftFailure} />
                         </td>
                         <td className="px-4 py-3 text-sm text-gray-600">
                           {formatDuration(session.durationSeconds)}

src/Web/autopilot-monitor-web/app/layout.tsx

@@ -80,7 +80,9 @@ export const metadata: Metadata = {
   icons: {
     icon: "/icon.svg",
     shortcut: "/icon.svg",
-    apple: "/icon.svg",
+    // Safari cannot use SVG for apple-touch-icon; must be an opaque PNG
+    // (served by the app/apple-icon.png file convention).
+    apple: "/apple-icon.png",
   },
 };
 

src/Web/autopilot-monitor-web/app/sessions/components/SessionInfoCard.tsx

@@ -59,7 +59,7 @@ export default function SessionInfoCard({ session, enrollmentDuration, displaySt
         <InfoItem label="Duration" value={enrollmentDuration ?? `${Math.round(session.durationSeconds / 60)} min`} tooltip={lastContactTooltip} />
         <InfoItem label="Events" value={session.eventCount.toString()} tooltip={lastContactTooltip} />
         <InfoItem label="Reboots" value={(session.rebootCount ?? 0).toString()} tooltip="System reboots observed during enrollment (V2 only)" />
-        <InfoItem label="Status" value={<StatusBadge status={displayStatus} failureReason={session.failureReason} failureSource={session.failureSource} adminMarkedAction={session.adminMarkedAction} reconcileReason={session.reconcileReason} />} />
+        <InfoItem label="Status" value={<StatusBadge status={displayStatus} failureReason={session.failureReason} failureSource={session.failureSource} adminMarkedAction={session.adminMarkedAction} reconcileReason={session.reconcileReason} espSoftFailure={session.espSoftFailure} />} />
         <InfoItem label="Enrollment Type" value={enrollmentTypeLabel(session, isGatherRulesSession)} />
         <InfoItem label="Join Type" value={joinTypeLabel(session)} />
         {validatedByLabel(session) && (
@@ -84,6 +84,19 @@ export default function SessionInfoCard({ session, enrollmentDuration, displaySt
           </div>
         </div>
       )}
+      {displayStatus === "Succeeded" && session.espSoftFailure && (
+        <div className="mt-4 flex items-start gap-2 px-3 py-2 rounded-lg bg-amber-50 border border-amber-200 text-sm text-amber-800 dark:bg-amber-900/30 dark:border-amber-800 dark:text-amber-200">
+          <svg className="w-4 h-4 flex-shrink-0 mt-0.5" fill="none" stroke="currentColor" viewBox="0 0 24 24">
+            <path strokeLinecap="round" strokeLinejoin="round" strokeWidth={2} d="M12 9v2m0 4h.01m-6.938 4h13.856c1.54 0 2.502-1.667 1.732-2.5L13.732 4c-.77-.833-1.964-.833-2.732 0L4.082 16.5c-.77.833.192 2.5 1.732 2.5z" />
+          </svg>
+          <span>
+            <strong>Completed with issues:</strong> the ESP reported a terminal failure, but the user
+            reached a working desktop — most likely via the &quot;Continue anyway&quot; button. At least one
+            blocking item did not finish inside the ESP window; check the timeline and app list for the
+            item that held the ESP at its timeout wall.
+          </span>
+        </div>
+      )}
       {ntpOffset && Math.abs(ntpOffset.offsetSeconds) > 30 && (
         <div className="mt-4 flex items-center gap-2 px-3 py-2 rounded-lg bg-amber-50 border border-amber-200 text-sm text-amber-800">
           <svg className="w-4 h-4 flex-shrink-0" fill="none" stroke="currentColor" viewBox="0 0 24 24">
@@ -250,12 +263,12 @@ function ReconcileTimingFacts({ session }: { session: Session }) {
   );
 }
 
-function StatusBadge({ status, failureReason, failureSource, adminMarkedAction, reconcileReason }: { status: string; failureReason?: string; failureSource?: string; adminMarkedAction?: string; reconcileReason?: string }) {
+function StatusBadge({ status, failureReason, failureSource, adminMarkedAction, reconcileReason, espSoftFailure }: { status: string; failureReason?: string; failureSource?: string; adminMarkedAction?: string; reconcileReason?: string; espSoftFailure?: boolean }) {
   // Delegate the status pill (+ timeout affordance + "manual"/"reconciled" badges) to the shared,
   // canonical SessionStatusBadge so this page picks up every status — incl. the AwaitingUser/Incomplete
   // reclassification states — from one source instead of a divergent local map.
   const badge = (
-    <SessionStatusBadge status={status} failureReason={failureReason} adminMarkedAction={adminMarkedAction} reconcileReason={reconcileReason} />
+    <SessionStatusBadge status={status} failureReason={failureReason} adminMarkedAction={adminMarkedAction} reconcileReason={reconcileReason} espSoftFailure={espSoftFailure} />
   );
 
   // Session-detail-only affordance: link a rule-attributed failure back to its analyze rule.

src/Web/autopilot-monitor-web/app/settings/tenant/sections/SectionOptionalGraphCapabilities.tsx

@@ -4,6 +4,7 @@ import { useCallback, useEffect, useState } from "react";
 import { useAuth } from "../../../../contexts/AuthContext";
 import { useTenant } from "../../../../contexts/TenantContext";
 import { api } from "@/lib/api";
+import { AGENT_DOWNLOAD_URL } from "@/utils/config";
 import { authenticatedFetch } from "@/lib/authenticatedFetch";
 import { trackEvent } from "@/lib/appInsights";
 
@@ -111,8 +112,9 @@ export function SectionOptionalGraphCapabilities() {
   const allPermissions = Array.from(new Set(featureRows.flatMap(f => f.requiredPermissions)));
   const allPermissionsLine = allPermissions.map(p => `"${p}"`).join(",");
 
+  const scriptDownloadUrl = `${AGENT_DOWNLOAD_URL}/Grant-AutopilotMonitorAddOn.ps1`;
   const psCommand = status?.clientId
-    ? `.\\Grant-AutopilotMonitorAddOn.ps1 \`\n    -ClientId "${status.clientId}" \`\n    -Permissions ${allPermissionsLine} \`\n    -TenantId "${tenantId ?? "<your-tenant-id>"}"`
+    ? `irm '${scriptDownloadUrl}' -OutFile .\\Grant-AutopilotMonitorAddOn.ps1\n.\\Grant-AutopilotMonitorAddOn.ps1 \`\n    -ClientId "${status.clientId}" \`\n    -Permissions ${allPermissionsLine} \`\n    -TenantId "${tenantId ?? "<your-tenant-id>"}"`
     : "";
 
   return (
@@ -205,8 +207,16 @@ export function SectionOptionalGraphCapabilities() {
         <div className="bg-white shadow rounded-lg p-6">
           <h3 className="text-md font-semibold text-gray-900 mb-2">PowerShell grant command</h3>
           <p className="text-sm text-gray-600 mb-3">
-            Open a Windows PowerShell or PowerShell 7 prompt as a tenant administrator and run the script
-            below. It is idempotent — re-running it skips permissions that are already granted.
+            Open a Windows PowerShell or PowerShell 7 prompt as a tenant administrator and run the commands
+            below — the first line downloads the script (
+            <a
+              href={scriptDownloadUrl}
+              className="text-green-700 underline hover:text-green-800"
+              download
+            >
+              direct download
+            </a>
+            ), the second runs it. It is idempotent — re-running it skips permissions that are already granted.
           </p>
 
           <pre className="bg-gray-900 text-gray-100 text-xs font-mono p-3 rounded overflow-x-auto">

src/Web/autopilot-monitor-web/components/SessionStatusBadge.tsx

@@ -5,7 +5,9 @@
  * failure gets a small ⏱️ affordance. When {@link adminMarkedAction} is set (an administrator manually flipped
  * the terminal state), a small "manual" badge is appended. When {@link reconcileReason} is set (the BACKEND
  * declared the success — timeout-sweep reconcile or late-completion upgrade), a small "reconciled" badge is
- * appended with the justification as tooltip — otherwise the bare pill is returned unchanged.
+ * appended with the justification as tooltip. When {@link espSoftFailure} is set (the session completed while
+ * an ESP-failure advisory was unresolved — Continue-Anyway observation / advisory defang), an amber
+ * "with issues" badge is appended — otherwise the bare pill is returned unchanged.
  */
 
 const STATUS_CONFIG: Record<string, { color: string; text: string }> = {
@@ -26,6 +28,7 @@ export function SessionStatusBadge({
   failureReason,
   adminMarkedAction,
   reconcileReason,
+  espSoftFailure,
 }: {
   status: string;
   failureReason?: string | null;
@@ -33,6 +36,8 @@ export function SessionStatusBadge({
   adminMarkedAction?: string | null;
   /** Set when the backend declared the success (sweep reconcile / late-completion upgrade) — appends a "reconciled" badge. */
   reconcileReason?: string | null;
+  /** Set when the session succeeded despite an unresolved ESP failure (Continue-Anyway) — appends an amber "with issues" badge. */
+  espSoftFailure?: boolean | null;
 }) {
   const config = STATUS_CONFIG[status] || STATUS_CONFIG.Unknown;
   // The ⏱️ affordance marks the silence/timeout family: a "timed out" Failed, or any Incomplete
@@ -54,8 +59,12 @@ export function SessionStatusBadge({
   // Backend-declared success: adminMarkedAction wins (the "manual" badge already attributes the
   // flip); a reconciled badge on top would double-attribute the same verdict.
   const showReconciled = !adminMarkedAction && status === "Succeeded" && !!reconcileReason;
+  // Soft failure: the device reached a working desktop, but the ESP gave up on at least one
+  // blocking item first (user pressed "Continue anyway"). Amber to match the semantic weight —
+  // enrolled and usable, yet not everything the ESP tracked finished inside its window.
+  const showSoftFailure = status === "Succeeded" && !!espSoftFailure;
 
-  if (!adminMarkedAction && !showReconciled) return pill;
+  if (!adminMarkedAction && !showReconciled && !showSoftFailure) return pill;
 
   return (
     <span className="inline-flex items-center gap-1.5">
@@ -67,13 +76,21 @@ export function SessionStatusBadge({
         >
           manual
         </span>
-      ) : (
+      ) : showReconciled ? (
         <span
           className="px-1.5 py-0.5 text-[10px] leading-4 font-semibold rounded border border-sky-300 bg-sky-50 text-sky-700 dark:border-sky-700 dark:bg-sky-900/40 dark:text-sky-300"
           title={reconcileReason || undefined}
         >
           reconciled
         </span>
+      ) : null}
+      {showSoftFailure && (
+        <span
+          className="px-1.5 py-0.5 text-[10px] leading-4 font-semibold rounded border border-amber-300 bg-amber-50 text-amber-700 dark:border-amber-700 dark:bg-amber-900/40 dark:text-amber-300"
+          title="ESP reported a terminal failure, but the user most likely pressed 'Continue anyway' and reached the desktop — at least one blocking item did not finish inside the ESP window"
+        >
+          with issues
+        </span>
       )}
     </span>
   );

src/Web/autopilot-monitor-web/lib/__tests__/rulePageHelpers.test.ts

@@ -0,0 +1,23 @@
+import { describe, it, expect } from "vitest";
+import { bumpVersion } from "../rulePageHelpers";
+
+describe("bumpVersion", () => {
+  it.each([
+    ["1.0", "1.1"],
+    ["1.9", "1.10"],
+    ["1.0.0", "1.1.0"], // three segments stay three segments
+    ["2.3.7", "2.4.0"], // patch resets on a minor bump
+    ["1.0.0.5", "1.1.0.0"],
+  ])("bumps %s to %s preserving the segment count", (input, expected) => {
+    expect(bumpVersion(input)).toBe(expected);
+  });
+
+  it("falls back to 1.1 for empty or missing input", () => {
+    expect(bumpVersion("")).toBe("1.1");
+    expect(bumpVersion(undefined as unknown as string)).toBe("1.1");
+  });
+
+  it("treats a non-numeric minor as 0", () => {
+    expect(bumpVersion("1.x")).toBe("1.1");
+  });
+});

src/Web/autopilot-monitor-web/lib/rulePageHelpers.ts

@@ -25,10 +25,16 @@ export function stripInternalFields<T extends { isBuiltIn?: any; isCommunity?: a
   return rest as Omit<T, "isBuiltIn" | "isCommunity" | "createdAt" | "updatedAt">;
 }
 
-/** Bump a semver-style version string: "1.0" -> "1.1", "1.9" -> "1.10", etc. */
+/**
+ * Bump a semver-style version string's minor segment, preserving the segment
+ * count and zeroing anything after the minor: "1.0" -> "1.1", "1.9" -> "1.10",
+ * "1.0.0" -> "1.1.0", "2.3.7" -> "2.4.0".
+ */
 export function bumpVersion(v: string): string {
-  const parts = (v ?? "1.0").split(".");
-  const major = parts[0] ?? "1";
-  const minor = parseInt(parts[1] ?? "0", 10);
-  return `${major}.${minor + 1}`;
+  const parts = (v || "1.0").split(".");
+  const major = parts[0] || "1";
+  const parsedMinor = parseInt(parts[1] ?? "0", 10);
+  const minor = Number.isNaN(parsedMinor) ? 0 : parsedMinor;
+  const trailingZeros = parts.slice(2).map(() => "0");
+  return [major, `${minor + 1}`, ...trailingZeros].join(".");
 }

src/Web/autopilot-monitor-web/types/session.ts

@@ -18,6 +18,10 @@ export interface Session {
   failureSource?: string;
   /** Non-empty only when the BACKEND declared this session Succeeded (timeout-sweep reconcile or late-completion upgrade) — carries the justification so operators can tell it apart from an agent-reported success. */
   reconcileReason?: string;
+  /** True only on a Succeeded session that completed while an ESP failure was unresolved — the user most likely pressed "Continue anyway". Rendered as the amber "with issues" badge. */
+  espSoftFailure?: boolean;
+  /** How a soft-failure completion was detected: "continue_anyway_observation" | "continue_anyway_post_accountsetup". Empty on clean successes. */
+  completionSource?: string;
   /** Non-null only when an administrator flipped the session manually via the portal. Values: "Succeeded" | "Failed". */
   adminMarkedAction?: string;
   /** Backend device-validation path that accepted the device at registration: "AutopilotV1" | "CorporateIdentifier" | "DeviceAssociation" | "Bootstrap" — absent for sessions before this feature or tenants with device validation off. */

src/Web/autopilot-monitor-web/utils/__tests__/guardValidation.test.ts

@@ -196,6 +196,32 @@ describe("validateWmiTarget", () => {
     const r = validateWmiTarget("", false);
     expect(r.allowed).toBe(false);
   });
+
+  // Property projections of an allowed class expose a strict subset of the
+  // already-allowed SELECT * and are therefore admitted.
+  it.each([
+    "SELECT BatteryStatus FROM Win32_Battery",
+    "SELECT BatteryStatus, EstimatedChargeRemaining FROM Win32_Battery",
+    "SELECT BatteryStatus,EstimatedChargeRemaining FROM Win32_Battery",
+    "select batterystatus from win32_battery",
+    "SELECT\tBatteryStatus\tFROM\tWin32_Battery",
+    "  SELECT BatteryStatus FROM Win32_Battery  ",
+    "SELECT Status FROM Win32_Battery WHERE BatteryStatus = 1",
+  ])("allows projection of an allowed class: %s", (query) => {
+    expect(validateWmiTarget(query, false).allowed).toBe(true);
+  });
+
+  it.each([
+    "SELECT Name FROM Win32_Process", // class not on the allowlist
+    "SELECT BatteryStatus FROM Win32_BatteryX", // class spoofing
+    "SELECT *, Name FROM Win32_BIOS", // star mixed into a property list
+    "SELECT FROM Win32_BIOS", // empty property list
+    "SELECT BatteryStatus FROM", // missing class
+    "SELECT BatteryStatus FROM Win32_Battery;DROP", // non-whitespace boundary
+    "SELECT BatteryStatus, FROM Win32_Battery", // trailing comma
+  ])("rejects malformed or disallowed projection: %s", (query) => {
+    expect(validateWmiTarget(query, false).allowed).toBe(false);
+  });
 });
 
 // ---------------------------------------------------------------------------

src/Web/autopilot-monitor-web/utils/config.ts

@@ -49,6 +49,13 @@ export const API_URL_PROD = "https://autopilotmonitor-api-eu.azurewebsites.net";
 /** Production blob origin (CSP connect-src: diagnostics SAS uploads). */
 export const BLOB_URL_PROD = "https://autopilotmonitoreu.blob.core.windows.net";
 
+/**
+ * Customer-facing download alias for agent artifacts and helper scripts
+ * (Front Door in front of the blob origin). Counterpart of
+ * Constants.AgentDownloadBaseUrl on the C# side.
+ */
+export const AGENT_DOWNLOAD_URL = "https://download.autopilotmonitor.com/agent";
+
 /**
  * Hostnames the portal accepts in a bootstrap response's agentDownloadUrl.
  * Keep in sync with ValidateBootstrapCodeFunction.cs, which builds that URL

src/Web/autopilot-monitor-web/utils/guardValidation.ts

@@ -39,6 +39,29 @@ const ALLOWED_COMMANDS_SET = new Set(
   ALLOWED_COMMANDS_LIST.map((c) => c.toLowerCase())
 );
 
+// Canonical WMI allowlist entry: exactly "SELECT * FROM <class>".
+// Such entries contribute their class name; any other entry (none today)
+// stays on the legacy literal-prefix path. Mirrors the agent's
+// GatherRuleGuards derivation.
+const CANONICAL_WMI_ENTRY_REGEX = /^SELECT\s+\*\s+FROM\s+([A-Za-z_][A-Za-z0-9_]*)$/i;
+
+// WQL shape "SELECT <* | property-list> FROM <class>", class token bounded by
+// whitespace or end of string (a trailing WHERE clause is permitted,
+// "Win32_BIOSX" cannot pass as "Win32_BIOS").
+const WMI_QUERY_REGEX =
+  /^SELECT\s+(?:\*|[A-Za-z_][A-Za-z0-9_]*(?:\s*,\s*[A-Za-z_][A-Za-z0-9_]*)*)\s+FROM\s+([A-Za-z_][A-Za-z0-9_]*)(?=\s|$)/i;
+
+const ALLOWED_WMI_CLASSES = new Set<string>();
+const RESIDUAL_WMI_PREFIXES: string[] = [];
+for (const entry of ALLOWED_WMI_QUERY_PREFIXES) {
+  const match = CANONICAL_WMI_ENTRY_REGEX.exec(entry.trim());
+  if (match) {
+    ALLOWED_WMI_CLASSES.add(match[1].toLowerCase());
+  } else {
+    RESIDUAL_WMI_PREFIXES.push(entry);
+  }
+}
+
 const BLOCKED_USERS_PREFIX = BLOCKED_FILE_PREFIXES[0] || "C:\\Users";
 
 // ---------------------------------------------------------------------------
@@ -267,7 +290,15 @@ export function validateWmiTarget(
     return { allowed: true, reason: "All WMI queries allowed in unrestricted mode", unrestricted: true };
   }
 
-  for (const prefix of ALLOWED_WMI_QUERY_PREFIXES) {
+  // "SELECT <* | property-list> FROM <class>" against an allowed class — a
+  // property projection exposes a strict subset of the already-allowed SELECT *.
+  const match = WMI_QUERY_REGEX.exec(trimmed);
+  if (match && ALLOWED_WMI_CLASSES.has(match[1].toLowerCase())) {
+    return { allowed: true, reason: `Class ${match[1]} is on the WMI allowlist`, unrestricted: false };
+  }
+
+  // Non-canonical allowlist entries keep the legacy literal-prefix match.
+  for (const prefix of RESIDUAL_WMI_PREFIXES) {
     if (trimmed.length >= prefix.length) {
       const candidate = trimmed.substring(0, prefix.length);
       if (candidate.toLowerCase() === prefix.toLowerCase()) {
@@ -280,7 +311,7 @@ export function validateWmiTarget(
 
   return {
     allowed: false,
-    reason: "Not a recognized WMI query prefix",
+    reason: "Not an allowed WMI query — use SELECT <properties or *> FROM <allowed class>",
     unrestricted: false,
   };
 }