LogParserCollector and EventLogCollector called no guard at all, while
File/Json/Xml did — and rules/gather-rules.md already promised customers
that log-parser paths were allowlisted. A tenant admin could author a
logparser rule against C:\Users via POST /api/rules/gather and have up to
1000 lines uploaded as telemetry, bypassing the hard C:\Users block the
same page guarantees; eventlog could read any channel, Security included.
The backend validates no collectorType, target, or parameters, so the
agent is the only place this can be enforced.
Both now guard. logparser cannot check rule.Target directly (GetFullPath
throws on a wildcard), so it checks the directory for wildcard targets and
re-checks each resolved path. Event log channels get an allowlist matched
on a "/" boundary. Hard-blocked channels live in a C# constant rather than
guardrails.json: a parse error drops the allowlists to empty, which is
fail-closed, but would have lifted a block sourced from the same file.
Also fixes an inert timeout in CommandCollector: ReadToEnd() blocks until
the pipe closes, so WaitForExit(30000) was only reached once the process
had already exited and a hung command pinned its worker indefinitely.
Adds the first real CI gate. No workflow ran tests on a pull request — the
deploy workflows build without testing and build-agent.yml covered only
DecisionCore.Tests on manual dispatch — while trust/security-faq.md
already told customers automated tests run in CI. ci.yml now runs all six
suites on every PR and merge, plus a job that fails on generated-guardrail
drift, alongside CodeQL and Dependabot. OfficeBinaryWatcherTests is marked
SerialThreading; its bounded re-probe races the shared ThreadPool and
failed once under full-suite load.
Verified: 3101 backend, 1911 agent V2, 474 decision core, 44 dialog, 572
web, 447 MCP. The timeout fix is verified by runtime, not assertion — a
Start-Sleep 120 command now returns after ~30s.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A German question returns zero results from search_docs, not poor ones: both the
documentation and the rules are English, and all-MiniLM-L6-v2 is an English-only
model, so the query vector lands nowhere near any chunk. For a German-speaking
admin driving an MCP client in their own language, the tool is effectively blind
and gives no hint why.
The caller is itself a language model, so the cheapest correct fix is to say so:
both tool descriptions now instruct it to translate the question to English before
searching and to answer the user in their own language. No infrastructure, no
model change, no ranking impact.
A multilingual embedding model would handle it natively and was considered, but it
is substantially larger — which matters on a scale-to-zero container — and it would
change English retrieval as well, so it would have to be measured against
scripts/eval-docs-search.ts rather than swapped in on the assumption that more
languages cannot hurt.
Verified: 447 tests pass, tsc clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The boot marks answered the question on the first production cold start, and the
answer overturns the reasoning behind an infrastructure change made earlier today.
+244ms node started (startup + module loading)
+431ms corpora read from disk (+187ms)
+491ms index read, validated, hydrated (+60ms for 3.12 MB)
+504ms listening
+677ms query embedder warm (background)
Against a 31.5s cold start, the application accounts for half a second. The other
~98% is Container Apps activation: KEDA reacting, pod scheduling, mounting the
236 MB image, starting the container. The container also boots faster than a
development laptop (504ms vs 1273ms), so nothing here is CPU-starved.
Consequences recorded rather than quietly dropped:
- Raising the container to 0.5 vCPU was justified by the embedding model load on
the cold-start path. Prebaked, that load is 173ms and runs in the background
where only the first search could wait on it. The justification does not survive
measurement. The setting stays — it plausibly helps concurrent requests, and at
scale-to-zero the cost difference is negligible — but the Bicep comment now says
what is true instead of what was assumed, so nobody re-derives the wrong reason.
- Index size is irrelevant to startup (60ms for 3.12 MB), so the corpus can grow
freely, and the embedding rounding earns its keep on file size alone.
- The only lever that would actually remove the cold start is minReplicas: 1, at
roughly 7x the monthly free consumption grant.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A cold start after scale-to-zero measures 24.5s to first byte against 60ms warm,
and nothing outside the process could say why. Log Analytics stamps every console
line of one ingestion batch with the same time, so the boot sequence collapses to
a single instant however it is queried, and piping stdout locally buffers until
exit. So the server times itself: `[boot +Nms]` marks derived from
process.uptime(), which unlike an in-process timer also covers Node's own startup
and module loading.
Measuring the precomputed path locally already contradicts two assumptions worth
recording:
+860ms node started (startup + module loading)
+1199ms corpora read from disk (+339ms)
+1242ms index read, validated, hydrated (+43ms for 3.12 MB)
+1273ms listening (+31ms)
+1497ms query embedder warm (+224ms, background)
The index is 43ms, not seconds — its size is not the cold-start cost, and growing
the corpus is cheap. The prebaked embedder is 224ms and loads in the background,
so only the first search could ever wait on it. That was the stated reason for
raising the container to 0.5 vCPU, which now looks like a weaker justification
than it did; the production marks will settle it. If they show the app listening
after a few seconds, the rest of the 24.5s is platform activation that neither CPU
nor index changes can touch.
Verified: 447 tests pass, marks appear on both boot paths locally. Incidentally
the run also confirmed the corpus hash guard and the staleness signal both work —
the local docs repo has moved to 6333d56 while the deployed image reports d8cb7f4,
so the precomputed index was correctly rejected and recomputed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The form offered trigger, phase scope and emit mode unconditionally, which
produced combinations that cannot do anything plus one silently wrong state:
- A scope on a Phase Start / Phase End rule that names a concrete phase can only
ever suppress the single firing — the phase IS the moment. Same for emit mode:
a one-shot has no second result to dedup against.
- Selecting "From a phase onwards" (or "Only during specific phases") and not
filling the detail dropdown serialised to null, i.e. the rule ran everywhere
while the form still read "From a phase onwards".
supportsPhaseScope / supportsEmitMode now drive visibility; buildScopeFields
nulls hidden-control state so a value from an earlier trigger choice cannot leak
into the payload; validateScopeSelection rejects an unfilled scope mode on both
create and save. Startup is the deliberate exception — a scope there defers the
one-shot instead of suppressing it, so the scope control stays and only emit mode
is hidden. One-shot triggers now state in-form when they collect.
Sub-dropdown hints no longer claim "No selection = all phases" (they turn amber
and name the required action instead).
Tests: app/gather-rules/__tests__/gatherRuleFormLogic.test.ts (15) pins the
applicability matrix, the payload nulling incl. the trigger-switch leak, the
unfilled-scope rejection, and the JSON-mode scopeMode derivation.
vitest 572/572; tsc silent.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The live test found a third recall gap. The fallback only topped up when semantic
search returned FEWER than topK results — but the more common failure is a full
list of unconvincing ones. Querying "WOW6432Node" returned three chunks scoring
~0.29, none containing the word, while the single chunk that does was never
considered at all.
The fix is a ranking change, so it was measured rather than reasoned about. Adds
scripts/eval-docs-search.ts, which scores strategies over 2465 queries in three
deliberately different classes: every chunk heading (easy by construction, so a
regression detector), rare identifiers (the class this fallback serves), and
hand-written questions (the reality anchor — not derived from the corpus, so it
shares little vocabulary with it). Embeddings are computed once and each strategy
is a pure function over the same candidates, so sweeping a threshold costs no
extra inference.
known-item literal handwritten MRR fixed broken
append-only (before) 71.9% 66.2% 1.000 - -
weak < 0.30 71.9% 72.6% 1.000 142 0
weak < 0.35 73.7% 81.1% 1.000 335 0
weak < 0.40 76.3% 88.2% 0.944 502 3
weak < 0.45 77.2% 93.0% 0.889 613 5
0.35 is strictly dominant — better in every class, worse in none — so that is
what ships. Higher thresholds keep repairing identifier queries but start
demoting hand-written questions, the class that most resembles real use, so the
gain stops being free. When the head is weak, exact matches move to the front
while at least one semantic result is kept, so a weak-but-correct hit is never
pushed out entirely.
The measurement also resized the problem: identifier queries were succeeding only
66.2% of the time, so the reported case was one of roughly 750, not an outlier.
That is the part a single anecdote could not have shown, and the reason the
harness exists before the fix rather than after it.
Verified: 447 tests pass, including new cases for the weak path and for leaving a
confident ranking untouched. On the real corpus "WOW6432Node" now returns the
chunk that contains it at rank 1, while "how do I deploy the agent with Intune"
(0.819) and "what firewall URLs do I need to allow" (0.374) are unchanged and the
negative controls stay empty.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two measured recall gaps in search_docs, both silent.
1. Truncation. MiniLM's context is 256 tokens; 87 of 324 chunks exceeded it and
13% of all tokens never reached the ranking — the tokenizer drops the tail
without complaint. Concretely, in built-in-rules.md :: Apps (734 tokens) the
strings ANALYZE-APP-013, ANALYZE-OFFICE-001, ANALYZE-CORR-003 and Click-to-Run
all sit past the cutoff, so "proxy configuration blocking app downloads" did not
retrieve that section at all.
Shrinking chunks cannot fix this: the corpus ranges 2.43 to 5.81 characters per
token, so no character budget bounds tokens, and a budget safe for the worst case
(~620 chars) would shred prose and tables alike. So the chunk stays whole and the
embedding is split instead — index() cuts long text into 250-token windows with
40 overlap, a document carries embeddings[][], and search() scores it by its BEST
window. Max rather than mean: averaging would blur a long section's several
topics into one mediocre direction, the same dilution that makes large context
windows worse than chunking.
The tokenizer is a build-time dependency only (vocabulary, not ONNX weights,
~93 ms); production boots from precomputed vectors and never loads it.
2. No general fallback. Proper nouns embed poorly — "RealmJoin" occurs in 3
chunks yet scores under the threshold. Fuzzy matching was built, measured and
rejected: at a threshold low enough to surface RealmJoin (0.28), Fuse also
returns 5 hits for "SCEPman", a word this corpus contains zero times. Literal
containment has neither problem, and every provider already exposes it. So a
short result set is now topped up via lexicalMatch on distinctive query terms,
requiring two matching terms as corroboration — without that rule "how do I bake
sourdough bread" matched a page about LLM providers, because "bake" is a
substring of "baked into the image". Keyword hits are appended and flagged, never
sorted together with cosine scores.
Also sharpens the tool description with the measured reason to prefer this over
fetching the docs site (~1k tokens for topK=3 against ~9k raw / ~600k rendered,
version-matched, and matching on meaning), and replaces the growing tail of
positional docs parameters with a DocsSearchBundle.
Verified: token coverage 87% -> 100%; "Office Click-to-Run install never
finished" 0.318 -> 0.442; "proxy configuration blocking app downloads" absent ->
0.559 at rank 1; the ten reference queries keep byte-identical scores and rank 1
(short chunks are one window, so the same vector as before — no long-chunk bias).
The fallback is not a dead path: across 904 rare corpus tokens queried alone it
contributes hits for 431, while negative controls stay empty. Precompute output
is accepted by the boot validator and a mutated corpus still rejected; index
2.63 -> 3.12 MB for 324 chunks / 424 vectors. The rules corpus gains the same
treatment unplanned: 128 rules now carry 172 vectors. 445 tests pass.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The MCP server could answer telemetry questions ("why did this session fail")
but not product ones ("how do I deploy the agent", "where is my data stored").
Those answers already exist in the autopilotmonitor-docs bundle; search_docs
makes them semantically searchable.
Deliberately a separate index from search_knowledge: rules and prose answer
different questions, and sharing one topK would dilute both. Registered for
every role — published documentation carries no tenant data — and skipped
entirely when no corpus was baked in, so a doc-less build advertises no
broken tool.
Corpus handling:
- Chunked at h2/h3 and <details>/<summary>, each chunk prefixed with a
"Page > Section > Subsection" breadcrumb so a mid-page hit is self-describing.
- Newlines are normalized before any parsing. This is a correctness
precondition, not cosmetics: the bundle ships CRLF and JS `.` does not match
\r, so /^(#{2,3})\s+(.*)$/ matched zero headings — silently, degrading every
page to unlabelled paragraph splits.
- Two size exceptions: tables are never split (a split table leaves `| --- |`
rows with no header), and unbroken changelog lists split at line level (left
whole, their tails fall outside the model window and become unsearchable).
- Short sections are kept — a one-line answer is often the best hit, and the
breadcrumb supplies the context a short body lacks.
Build and delivery:
- Docs are embedded at image build time into an optional third
PrecomputedSection. Optional so a checkout without the docs repo does not
invalidate the rules and event-type vectors; the reverse (server has docs,
file does not) is staleness and is rejected.
- The deploy workflow checks the public docs repo into the build context and
passes its SHA as DOCS_COMMIT, which /health reports alongside the chunk
count — editing the docs does not update a running server, so staleness is
made checkable rather than memorable.
- Embeddings are rounded to 6 decimals on serialize: 55% smaller index (2.63
vs 5.88 MB) for a ~1e-5 norm error against a 1e-3 tolerance. indexPrecomputed
now asserts unit norm, since the hydrate path never checked it.
Infra: the Bicep template was two changes behind the live deployment (EU
cutover and ACR hardening, both applied directly in Azure). Reconciled against
the running resources — managed-identity image pull replaces the admin
credentials and the acr-password secret — and infra/ is no longer gitignored so
it stops drifting. Container raised to 0.5 vCPU / 1.0 GiB for the cold-start
query-embedder load. The template does not model the mcp.autopilotmonitor.com
custom domain, so a full deployment would remove it; the header warns about it.
Verified: 428 tests pass; 45 pages produce 324 chunks; sample queries hit the
right page top-1/2; all 8 derived URLs return 200 live; the precompute output
is accepted by the boot validator and a mutated corpus is rejected.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow-up to gather-rule phase scoping. One-shot collection at a phase START
already existed (phase_change dedups per (rule, phase)) and at enrollment END
(on_event + enrollment_complete); the END of a specific phase had no trigger.
New trigger value "phase_exit": fires once when TriggerPhase is LEFT (empty =
every phase transition). Together with phase_change it forms the two one-shot
bookends of a phase.
Agent (GatherRuleExecutor.OnPhaseChanged):
- Exit rules are evaluated BEFORE _currentPhase advances, so both the
TriggerPhase match and the phase-scope gate see the phase being LEFT. A rule
scoped to DeviceSetup therefore fires on its own exit even though the phase
being entered is out of scope.
- A transition into Failed fires the exit rules of the failing phase — the
snapshot at a failure boundary is the point of the trigger.
- Exit of Unknown never fires; a repeated identical phase signal is not a
transition.
- Dedup key "exit:{ruleId}|{phase}" shares _phaseRulesExecuted with the
phase_change keys but stays in a disjoint key space, so an enter-rule and an
exit-rule on the same phase cannot consume each other's slot.
Portal: trigger labels are now "Phase Start" / "Phase End"; the canonical phase
dropdown and the create/save payloads cover both via PHASE_TRIGGERS; the card
renders "entering X" / "leaving X".
No ConfigVersion bump: no new field is transported. Old agents do not know the
value and simply never fire the rule (silent no-op).
Tests: GatherRuleExecutorPhaseExitTests (11) — enter-vs-exit, no refire on
re-entry, empty-triggerPhase, exit-of-Unknown, same-phase-signal, into-Failed,
Complete, scope-against-left-phase, out-of-scope slot preservation, key-space
disjointness, on_change interplay. Build 0 Errors; V2.Core.Tests 1884/1884;
Functions.Tests 3099/3099; vitest 557/557; tsc silent.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Rework after the first real enrollment (session b2e890c1) exposed three
defects in a654c730:
- Shell-Core "RebootCoalescing" match was a false positive by construction:
the token appears in the ROUTINE SubcategoryProcessing_Started bootstrap
marker ("Starting subcategory DeviceSetup.RebootCoalescing...") on EVERY
enrollment. esp_reboot_coalescing removed entirely (branch, constant,
catalog, tests).
- 2800 reboot-required flags fire at every policy sync; only DeviceSetup-time
flags cause the coalesced reboot. The session's flags fired during
AccountSetup - no reboot - yet our Warning events claimed "device will
restart". Raw event is now a neutral Info observation; the reboot CLAIM
moved to ANALYZE-ESP-005 with a REQUIRED system_reboot_detected >= 1
condition (no observed reboot -> advisory silent).
- One event per 2800 record was spam: records are now buffered and debounced
(10 s) into ONE aggregated event per burst (rebootUris[]/uriCount/
firstRebootUri/recordCount/unparsedCount). Watermark is persisted at FLUSH,
not claim - a pre-emit reboot kill loses nothing, the post-restart backfill
re-emits the burst at its historical time.
Verified 2800 description shape recorded in code, tests, and concept doc:
"The following URI has triggered a reboot: (./Device/...)."; extraction
covered by a real-text regression test. Lesson added to tasks/lessons.md
(observation-level events vs outcome-gated claims; substring tokens appear
in framework bookkeeping records).
Tests: Agent.V2.Core 1867/1867, Functions 3099/3099, MCP 400 passed,
solution build 0 errors.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Device-assigned MDM policies whose CSP URIs match the OS reboot-required
catalog force a coalesced reboot at the end of ESP DeviceSetup - the user
sees an unexpected restart and a second sign-in before AccountSetup.
Invisible in the Intune console; admins simply lived with it. Now attributed
and actionable (PMPC research).
- MdmRebootPolicyTracker (+host): watches DeviceManagement-Enterprise-
Diagnostics-Provider/Admin EventID 2800, emits one mdm_policy_reboot_required
per reboot-forcing URI (Warning, ImmediateUpload; tolerant EventData-first
URI extraction with regex fallback; fail-soft OMADM RebootRequired probe).
Cross-restart RecordId watermark (mdm-reboot-watermark.json) - mandatory,
the coalesced reboot restarts the agent and the backfill re-reads the log.
- ShellCoreTracker: RebootCoalescing branch on the already-watched 62407
stream -> esp_reboot_coalescing corroboration event; deliberate deviation:
emits on backfill too (live emit races the reboot kill).
- ANALYZE-ESP-005 advisory (warning): names the first URI via {{rebootUri}},
remediation = reassign profiles device -> user groups; +20 confidence on
coalescing corroboration, +10 on >=2 URIs.
- Config: Collectors.MdmRebootPolicyWatcherEnabled (default ON, internal
steering only - agent-native collector, NOT an IAgentAnalyzer, no admin UI),
MdmRebootPolicyBackfillLookbackMinutes=60; ConfigVersion 35.
- Event types deliberately distinct from system_reboot_detected so the
backend RebootCount (exact-string match, 2 sites) stays uninflated.
- Docs: docs/agent/mdm-reboot-coalescing.md (+index/log). Exact 2800/62407
message texts are unverified until E2E VM validation (tolerant matching).
Tests: Agent.V2.Core 1868/1868 (21 new), Functions 3099/3099, MCP 400 passed
(event-types drift), solution build 0 errors.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The 2026-07-18 storage migration reset every row's system Timestamp,
exposing that AuditLogs, OpsEvents and Events displayed, filtered and
retention-swept on it (custom "Timestamp" writes were silent no-ops —
reserved system property). All migrated rows read as 2026-07-18.
- New Helpers/BusinessTimestamp: RowKey decoders (audit !revtick_guid,
ops bare revtick, event yyyyMMddHHmmssfff_seq), RowKey-range OData
clause builders (tick-exact, with the mandatory RowKey lt '"' guard
so no lower bound can ever match legacy bare-GUID audit rows), and
dual DateTimeOffset/DateTime column reads.
- Writes in all three tables now persist OccurredUtc (custom column,
copied verbatim by future migrations); no-op ["Timestamp"] removed.
- Read mappers resolve OccurredUtc -> RowKey decode -> system Timestamp;
GetEarliestSessionEventTimestampAsync uses the same resolution so
StartedAt/Duration reconciliation can no longer absorb the migration
date on migrated sessions.
- Date-window filters and retention moved off the system Timestamp onto
RowKey ranges; ops retention loses its full-table scan and the
post-migration mass-delete cliff; audit retention runs a temporary
second pass for legacy GUID rows (TODO: remove after 2027-03-01).
- New GlobalAdmin backfill endpoint POST /api/maintenance/
backfill-occurred-utc (dry-run by default, batched, idempotent MERGE
of only OccurredUtc, table=audit|ops) — Events rows are deliberately
never mutated; their read path decodes the RowKey instead.
- Tests: decoders, clause boundary proofs incl. GUID-safety pins,
mapper precedence, backfill row decisions (Functions.Tests 3078).
- Docs: docs/backend/business-timestamps.md (system Timestamp is never
authoritative; migration checklist rule).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
New rules/ section in the knowledge bundle: semantics table, from-phase latch
behavior (never via Failed, sticky through it), on_change x emitOnlyIfExists
composition, restart re-emit, no-phase-signal caveat, old-agent degradation,
and the prerequisite toggle-wipe fix. Index + log updated.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Re-home agents to a new API base URL through the agent/config response
(backend move or per-tenant EU->US region move) - the lesson from the
2026-07-18 WEU->GWC cutover, where the compiled-in API hostname hard-cut
every in-flight agent (no DNS indirection possible: mTLS breaks behind a
proxy, Flex custom-domain TLS is broken).
Backend (ConfigVersion 33):
- AdminConfiguration.AgentMigrateApiBaseUrl (global) +
AgentMigrateTenantOverridesJson (tenantId->URL; empty value pins a
tenant against the global target; override wins)
- GetAgentConfigFunction.ResolveMigrateTarget serves the validated target;
AgentMigrateServed / AgentMigrateRejected warnings as delivery evidence
- TableConfigRepository Store+Map; admin portal Global Settings round-trip
Agent (kill-switch-mirrored hardening):
- Shared AgentEndpointMigrationRules: https-only, *.azurewebsites.net on a
label boundary, no path/port/userinfo - enforced on BOTH sides
- Live fetch only: RemoteConfigService cache strips the field on write AND
read; EndpointMigration.ResolveTarget re-checks the fetch outcome
- AgentRuntimeConfig.Resolve returns before merge/integrity/bootstrap
cleanup on a pending migration (no spurious IntegrityCheckFailed against
a stale migration stub), Program.RunAgent rebuilds Phase 3+4 against the
new URL - single hop, kill signal wins before and after the hop
Tests: AgentEndpointMigrationTests (backend rules + resolution),
EndpointMigrationTests + cache-hardening strip (agent). Full suites green
(Functions 2992, Agent Core 1822, vitest 557, tsc silent).
Doc: docs/agent/endpoint-migration.md
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
gh attestation verify only passes with the repo's current name
(okieselbach/AutopilotMonitor); the signed workflow identity carries it,
and the old dashed name fails identity matching even though git-remote
redirects still work. Verified against the first attested dry-run ZIP.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Build numbers now come from a public counter blob (buildcounter-v2.txt,
reserve-before-build with ETag-CAS) shared by local build.ps1 and CI, so the
two paths can never mint the same number. The CI step that committed
buildcounter.txt bumps to main is gone; the repo file is now a gitignored
offline fallback for -SkipUpload smoke builds only.
CI parity fixes vs the local build:
- ship SummaryDialog.exe.config in the ZIP (WPF AppContextSwitchOverrides;
without it the dialog dies with 0x8007045A when spawned from SYSTEM)
- verify EXE version == manifest == reserved build number before zipping
- update the AdminConfiguration hash-oracle only on publish_as_stable
CI-built release ZIPs now carry a GitHub Artifact Attestation (Sigstore
keyless provenance, actions/attest-build-provenance@v4).
Doc: docs/agent/build-counter-blob.md
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Probe enrollment e7dbac7a (24H2) showed Windows writes IsAutoPilotDisabled
(capital P) while the troubleshooting FAQ documents IsAutopilotDisabled —
the missing-event flat evidence lookup was case-sensitive and would have
missed it. Registry value names are case-insensitive by contract, so the
dump dictionary now uses OrdinalIgnoreCase; original casing is preserved
in the emitted payload. Doc notes the casing caveat.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
In the normal IME deployment model the agent installs only after MDM
enrollment, so TPM attestation and ZTD errors from earlier (retried)
attempts sat invisible in the event log. The tracker now replays
Autopilot-channel Level 1-2 records from the last 4h at startup through
the regular emission pipeline.
Spam containment (answering "spam or backfilled event?" — individual
events, but marked and bounded):
- errors only (Warning-level 100 heartbeat excluded by XPath)
- harmless-ID downgrade + burst rollup still applies to backfilled records
- cap of 25 records per scan, newest first (tail logged, not emitted)
- RecordId watermark persisted to autopilot-error-backfill.json, advanced
by both backfill and live watcher -> restarts never re-emit
- events carry backfilled=true + original timeCreated; the event's own
timestamp is emission time, so backend StartedAt alignment is not
dragged backwards by pre-enrollment records
Docs: Observability matrix updated (TPM row now visible-via-backfill) +
new "Startup error backfill" section. Tests: 5 new (XPath, watermark
roundtrip/monotonicity, backfilled/live markers, rollup-on-backfill);
full V2 suite 1804 green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The agent is normally installed by an IME platform script AFTER MDM
enrollment succeeded, so pre-enrollment failures (enrollment blockers,
TPM attestation) are invisible in the normal flow — no enrollment means
no agent and no authenticable session. Await-enrollment mode does not
help either: the binary only polls for the MDM certificate and cannot
authenticate without it. Only bootstrap-token deployments (one pilot
customer) can report pre-MDM.
- New Observability section in docs/agent/autopilot-ztd-diagnostics.md
with a per-rule visibility matrix (normal / bootstrap / await) and the
follow-up option of backfilling Autopilot-channel error records
- ANALYZE-ENRL-003/004 descriptions state the limitation explicitly so
portal users understand why these rules rarely fire
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A flip-flopping user-scoped WHfB CSP read "disabled" once, the arm-C
completion synthesized HelloOutcome="Skipped" and the session completed
5s later while the user sat inside the Hello wizard that had started
230ms earlier (Shell-Core 62404, previously InformationalEvent-only).
Layer 1 — engine (DecisionCore):
- New DecisionSignalKind.HelloWizardStarted + set-once fact
DecisionState.HelloWizardStartedUtc (additive-nullable, census +
snapshot wired).
- Prevention: centralized HelloSatisfiedForCompletion /
HelloPolicyDisabledWithoutWizard predicates veto the policy-disabled
stand-in at all five Skipped-synthesis sites and in the
completion_waiting prerequisites; the AdvisoryCompletion handler gets
a HelloWizardObservedPromote branch (its fallthrough is
re-arm-or-fail).
- Cure: HandleHelloWizardStartedV1 retracts an engine-synthesized
"Skipped" (exact-case discriminator vs the lowercase tracker
vocabulary; first fact-nulling reducer, stage-agnostic because the
RealmJoin gate-defer parks synthetic skips pre-Finalizing), cancels
finalizing_grace, returns to AwaitingHello and arms HelloSafety.
- Agent path: ShellCoreTracker raises HelloWizardStarted before
FinalizingSetupPhaseTriggered; 62404 included in the startup backfill
(fire-once); coordinator timestamp mirror; adapter emit (fire-once).
Layer 2 — HelloTracker confirmation read:
- ApplyPolicyRead state machine: "disabled" commits only after a second
confirming read (~10s poll); "enabled" commits immediately;
disabled→enabled flags flipFlopDetected=true; pending keeps the
policy-unknown wait cadence. Audit fields confirmedReads /
flipFlopDetected on hello_policy_detected.
Fixtures: signal-kinds/hello-wizard-started-v1.json +
enrollment-sessions/userdriven-hello-wizard-unskip-v1.jsonl (772fe502
ordering, ends Completed with HelloOutcome=completed).
Tests: DecisionCore 462/462, V2.Core 1775/1775.
Docs: OKF concept doc docs/agent/hello-wizard-unskip.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
GitHub resolves /-prefixed links against the repo root (missing /docs),
so cross-doc navigation broke. Switched all agent docs to standard
relative links and updated the bundle convention in index.md + CLAUDE.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Adopt the Open Knowledge Format (GoogleCloudPlatform/knowledge-catalog)
for contributor/AI-facing technical docs: YAML frontmatter on the three
existing concept documents, new index.md (progressive disclosure, root
carries okf_version) and log.md, plus the maintenance convention in
.claude/CLAUDE.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Reorganize docs/ tree:
- docs/design/mcp-oauth-flow.md -> docs/mcp-oauth-flow.md (no customer data)
- enrollment-status-reclassification{,-plan}.md -> tasks/ (git-ignored work
files; they contained a customer domain that must not ship in the public repo)
- remove now-empty docs/design/ folder
Repoint the 22 source files whose comments referenced
docs/design/enrollment-status-reclassification.md to the new tasks/ path.
Comment/docstring text only; no functional code changed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Sequence diagram and explanation of the dual-identity connect flow (AI-product
account vs. Entra work account), including the frequently-missed requirement
that the default browser must hold an active vendor session (claude.ai) for
the callback in step 6 to be accepted.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Follow-up to 049b6e5b: tenants running with Windows Hello disabled AND
the User ESP skipped can never produce a Hello terminal event, so the
reconcile rule would never fire for them and silent post-desktop
sessions stayed mislabeled AwaitingUser -> Incomplete. Mirror the
agent's Hello-disabled fast-path (HandleDesktopArrivedV1: policy
disabled + SkipUserEsp + desktop => complete):
- ExtractRollup reads hello_policy_detected Data.helloEnabled and
esp_config_detected Data.skipUserStatusPage (contradicting
observations resolve pessimistically to enabled/required).
- Rule 4 accepts desktop + (Hello terminal OR policy-disabled +
SkipUserEsp). Hello disabled WITHOUT SkipUserEsp keeps waiting —
the agent's strong post-AccountSetup gate (session 08c99638) blocks
its fast-path there too, and rule 2 covers the rollup-complete case.
- esp_exiting stays rejected as success evidence (page-lifecycle event,
fires with errorCode=0 while apps are unfinished — lesson 08c99638).
- Snapshot v3 + Web block additionally carry skipUserEsp.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The timeout classifier only read the ESP registry rollup, so a session
where the user reached the desktop and completed Windows Hello — with
only the RealmJoin completion gate still pending when the agent went
silent — was labeled AwaitingUser (and would have graduated to
Incomplete at grace expiry). Desktop arrival + a positive Hello terminal
are the agent's own Classic completion AND-gate; the RJ gate
self-releases after 60 min, so such a session's enrollment succeeded and
only the completion report never left the device.
- EnrollmentTimeoutClassifier: rollup now carries DesktopArrived,
HelloResolved (hello_provisioning_completed/hello_skipped only) and
RealmJoinDetected/Resolved (realmjoin_resolved/realmjoin_timeout);
new rule 4 reconciles desktop+Hello sessions to Succeeded with an
RJ-aware reason. Explicit failure and agent_emergency_break still win.
- FailureSnapshotBuilder: schema v3 embeds the new evidence
(helloResolved, realmJoinDetected, realmJoinResolved).
- Web FailureSnapshotBlock: renders Hello completed + RealmJoin rows.
- Design doc: decision table updated; desktop_arrived ALONE stays
rejected as a completion signal.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The 2026-07-08 timeout reclassification (Incomplete = terminal, non-failure)
migrated SessionStats, FleetHealthStats and SessionStatusBuckets, but two
reporting surfaces were overlooked: the live UsageMetricsService.SessionMetrics
(both platform + tenant paths, feeding get_usage_metrics and the usage
dashboards) and the persisted daily UsageMetricsSnapshot. Terminal Incomplete
sessions were counted in Total but in no breakdown bucket, so
Succeeded+Failed+InProgress no longer reconciled to Total and operators lost
sight of those devices.
- SessionMetrics + UsageMetricsSnapshot gain an Incomplete / SessionsIncomplete
count, mirroring the migrated peers (Incomplete only, not AwaitingUser).
- Snapshot column stored + mapped in TableStorageService.Metrics; legacy rows
deserialize to 0 (no seed bump needed).
- Success rate unchanged everywhere: denominator was already Succeeded+Failed,
correctly excluding Incomplete/AwaitingUser. SLA paths verified already-correct
(BuildSnapshot + breach evaluator receive terminal-filtered sessions).
- Web: Incomplete card added to both usage views (slate, matches the badge).
- MCP: query_raw_sessions status enum learns AwaitingUser/Incomplete so callers
can filter by the new states.
- Test: UsageMetricsServiceWindowTests pins the Incomplete bucket + 50% rate on
the crcins 1/1/8 shape.
Backend build clean, full Functions suite 2825 pass, web + MCP tsc clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Addresses four review rounds on the enrollment-status reclassification
(timeout != failure; new AwaitingUser/Incomplete states).
Backend correctness:
- UpdateSessionStatusAsync ETag force-fallback now mirrors the normal path:
uses IsTerminalTransitionAllowed + admin-marked guard, stamps CompletedAt
for Incomplete, and persists FailureReason/FailureSnapshotJson for
Incomplete/AwaitingUser (not just Failed) so a contended write cannot
regress a terminal or drop the classification reason/snapshot.
- StoreSessionAsync CAS loop re-reads the terminal tuple in the fresh-read
window and preserves a just-landed terminal (Succeeded/Failed/Incomplete)
verbatim, so re-registration can no longer revert it via the Replace.
- SessionGraceHours + AbsoluteMaxSessionHours now roundtrip through the
tenant-config Store/Map pair (overrides were silently lost).
- ResolveGraceHours clamps the assumed agent cap to >= the real runtime
default (48h) until the agent honors the override, so a low override can
never terminalize a still-running session early.
- Incomplete now funnels through the terminal counter-reconcile seam, and
the maintenance sweep re-reconciles after the synthetic session_timeout
event (also fixes the pre-existing off-by-one on the Failed path).
Web:
- SessionInfoCard delegates its status pill to the shared SessionStatusBadge
(removes a divergent local map; picks up AwaitingUser/Incomplete).
- SessionTable search color map + status filter chips learn the new states.
- useSessionEvents uses isTerminalStatus() instead of a hard-coded
Succeeded/Failed check (no needless refetch on Incomplete).
- DataManagementSection copy no longer says "Failed - Timed Out".
Docs/comments:
- Plan doc + TenantConfiguration comments corrected (+3h/51h buffer, and the
timeout now reclassifies to AwaitingUser/Incomplete, not "Failed").
Tests: +ResolveGraceHours clamp case, +grace/abs-cap config roundtrip +
legacy-row defaults. Full backend suite green (2712), web tsc green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Make the agent's silent 48h absolute session-age break visible: on break, before
cleanup/exit, fire a SessionAgeEmergencyBreak AgentErrorReport over the resilient
emergency channel; the backend (Part B) materializes it as an agent_emergency_break
timeline event and the timeout classifier terminalizes the session instead of
waiting out grace.
Decoupled so the guard stays pure and its unit tests are untouched:
- CheckSessionAgeEmergencyBreak gains optional Action onBreakFired = null, invoked
once when the break is confirmed, before marker/cleanup, fully swallowed.
- New EmergencyBreakReporter.TrySend builds a throwaway auth bundle
(BackendClientFactory) and sends, bounded to a 5s wait. AgentBootstrap wires it;
GetAgentVersion made internal.
NOTE: agent targets net48 and was NOT compiled in this environment (only the .NET 8
SDK is available) — to be built/verified locally. The Shared AgentErrorType value is
verified (compiled in the backend build).
- SessionStatusBuckets: explicit AwaitingUser + Incomplete buckets (out of Other),
still reconciling to Total.
- Metrics summary: expose awaitingUser + incomplete; failure rate is now
Failed/(Succeeded+Failed) — Incomplete and non-terminal states are out of the
denominator (crcins drops from failed/total to the real terminal rate).
- SessionStats.IncompleteLastNDays; dashboard SuccessRatePct was already terminal-only.
- Daily aggregation already used Succeeded+Failed denominator — unchanged.
- Tests: SessionStatusBucketsTests. Suite: 2705 pass. Fleet Health Incomplete surfacing
deferred to PR5 (web) with its documented over-all-sessions semantics.
A device that reaches the desktop is enrolled regardless of an earlier sweep
timeout guess, so a genuine completion must be able to heal a prior verdict.
- Replace the terminal idempotency guard with pure IsTerminalTransitionAllowed:
Succeeded UPGRADES any non-Succeeded state (Failed/Incomplete/AwaitingUser →
Succeeded); Failed/Incomplete and AwaitingUser never overwrite an existing
terminal. Ingest already calls UpdateSessionStatusAsync(Succeeded) on a late
enrollment_complete without a pre-terminal skip, so the relaxed guard wires the
reconcile end-to-end.
- Clear stale FailureReason + FailureSnapshotJson on Succeeded (reconcile hygiene).
- Preserve admin-marked terminals: a late completion never auto-overrides a
hand-marked Failed/Incomplete.
- Tests: SessionStatusTransitionTests (full matrix, 20). Suite: 2705 pass.
ReportAgentErrorFunction now injects ISessionRepository and, on a
SessionAgeEmergencyBreak report, writes an agent_emergency_break EnrollmentEvent
into the session stream (Sequence one past the last event, idempotent, best-effort
— never turns the always-200 emergency channel into a retry loop). This is what
makes the agent's otherwise-silent 48h break visible in the timeline and lets the
timeout classifier see it. Pure BuildAgentEmergencyBreakEvent helper (Warning,
non-terminal) + tests. Functions.Tests: 2685 pass.
A completion can only arrive <= the agent's absolute cap (it self-destructs and
sends nothing after), so the grace buffer needn't wait hours for stragglers — it
only has to span one 2h maintenance-sweep cycle + minor clock skew, and reconcile
heals any premature Incomplete. Buffer 12→3h, default grace 60→51h.
The 5h→terminal grace was a magic 72h. Anchor it to the agent instead: the 6h
per-run brake emits enrollment_failed (→ Failed at ~6h), but the 48h absolute
emergency break (AbsoluteMaxSessionHours) is SILENT to the backend — it writes a
local marker, cleans up and exits with no terminal event. So a session only stays
non-terminal past 48h when the agent hit that silent cap and is provably gone.
Grace must therefore be >= the agent absolute cap (never mark Incomplete while the
agent could still legitimately enroll) and only slightly beyond it:
effectiveGrace = max(AbsoluteMaxSessionHours + 12h buffer, SessionGraceHours) = 60h default.
- EnrollmentTimeoutClassifier.ResolveGraceHours (pure, tested): floors grace at
agentCap + buffer; a SessionGraceHours override can only raise it.
- TenantConfiguration.SessionGraceHours default 0 (= auto-derive); mirror
AbsoluteMaxSessionHours so the floor follows a tenant override of the agent cap.
- Sweep uses ResolveGraceHours instead of the hard-coded 72h.
Follow-up noted: wire AbsoluteMaxSessionHours down into AgentConfigResponse so the
agent and backend share one value. Functions.Tests: 2678 pass.
Wire EnrollmentTimeoutClassifier into MarkStalledSessionsAsTimedOutAsync so the
5h inactivity sweep stops blindly failing silent-but-provisioned sessions
(docs/design/enrollment-status-reclassification.md). Forward-only.
- TenantConfiguration.SessionGraceHours (default 72h) + default config.
- Sweep Stage 2: classify from the ESP subcategory rollup instead of hard Failed.
Fast-path skips within-grace AwaitingUser sessions with NO event read (grace is
time-based; late completions are caught by ingest). Per-outcome counters; audit
+ ops event only when the pass actually terminalized/transitioned something.
- session_timeout event + auto-analyze kept ONLY for Failed/Incomplete (parity with
the agent max-lifetime path); AwaitingUser is non-terminal, Succeeded is a reconcile.
- GetStalledSessionsAsync includes AwaitingUser so it graduates to Incomplete once
grace elapses; agent-silent (Stage 1) stays InProgress-only.
- UpdateSessionStatusAsync: Incomplete is terminal (idempotency); AwaitingUser can't
regress a terminal; CompletedAt stamped for Incomplete (no skewing duration); reason
+ snapshot persisted for Incomplete/AwaitingUser.
Full Functions.Tests suite: 2670 pass.
Foundation for reclassifying 5h-timeout sessions (docs/design/
enrollment-status-reclassification.md). No behaviour change to the sweep yet —
this only adds the pieces PR2 wires in.
- SessionStatus: append AwaitingUser (non-terminal) + Incomplete (terminal,
non-failure). Members appended so persisted ordinals never shift.
- EnrollmentTimeoutClassifier: pure ExtractRollup (parses the esp_provisioning_status
"N of M subcategories completed" message + the 30s all-succeeded fallback line —
categorySucceeded is ignored, Windows never sets it) and ClassifyTimedOutSession
implementing the decision table (explicit failure→Failed; AccountSetup all-succeeded
/ enrollment_complete→Succeeded; DeviceSetup done + within grace→AwaitingUser, else
Incomplete; silent pre-DeviceSetup→Incomplete). Never returns Failed without an
explicit failure signal.
- FailureSnapshotBuilder: embed the same rollup (schema v2) so the snapshot records the
evidence behind the verdict.
- Tests: EnrollmentTimeoutClassifierTests (12). Full Functions.Tests suite 2670 pass.
Design note + implementation plan for splitting the 5h maintenance-sweep
"Failed" bucket into three states. Evidence (crcins.com, 90d): 90.7% of
"failures" are 5h inactivity timeouts with no explicit failure event; 99.1%
of those had DeviceSetup fully provisioned (device enrolled) and simply had
the agent go silent during the user/Account-Setup phase. True failure rate
~4-5%, not 57.2%.
Proposal (backend + stats only, no agent change / no heartbeat):
- add AwaitingUser (non-terminal) + Incomplete (terminal, non-failure) states
- classify timed-out sessions from the ESP subcategory rollup the agent
already emits (gate on subcategories, not the unreliable categorySucceeded)
- passive grace window owned by the backend; reconcile late completions to
Succeeded; exclude Incomplete from the failure-rate denominator
The detailed per-component chapters (Agent internals through Build & Development)
are retired — product documentation lives at docs.autopilotmonitor.com now, and
the deep internals went stale faster than they earned their keep. What remains
is corrected against the actual tree: Shared/DecisionCore and SummaryDialog.Tests
added, src/McpServer and rules/guardrails.json were missing entirely, scripts/
gained Backup + CustomerSetup, Functions references DecisionCore too, and the
overview diagram no longer claims the web app serves docs.
docs/images stays — all five files are embedded in the repo README.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The V1 agent line is retired: no new V1 sessions since 2026-05-09, only
orphaned straggler traffic remained on /api/agent/ingest (last seen
2026-06-29, events for long-cleaned-up sessions). Bootstrap/Intune/CI
have deployed V2-only since the v2 cutover.
Agent:
- Delete AutopilotMonitor.Agent, .Agent.Core, .Agent.Core.Tests
(projects + sln entries). V2 trio, SummaryDialog, DecisionCore.Tests
and V2Isolation.targets stay.
Backend:
- Relocate the shared statics/DTOs V2 reused from the V1 partials
(StampServerFields, SanitizeEventTimestamps, IsMaxLifetimeAgentShutdown,
RecalculateAppDurations, EventClassification, AppInstallAggregationState)
into EventIngestProcessor; IngestEventsOutput moves next to
IngestTelemetryFunction (same namespace, zero ripple).
- Delete IngestEventsFunction (route agent/ingest), BootstrapIngestEvents
(route bootstrap/ingest), NdjsonParser and their DI registration.
EventDataNormalizer stays (used by TelemetryPayloadParser).
- Drop both routes from EndpointAccessPolicyCatalog and
RequestTelemetryMiddleware.TenantHeaderTrustedPaths.
- Retire the V1 update-hash oracle: LatestAgentV1*/LatestBootstrapV1*
fields + GetAgentLine case 1 (default arm returns empty hashes, so
stragglers gracefully skip the integrity check; ParseAgentMajor keeps
its default of 1 on purpose — defaulting to 2 would feed V2 hashes to
V1 binaries). TableConfigRepository Store+Map updated in lockstep.
- Remove ApiEndpoints.IngestEvents/.BootstrapIngestEvents constants.
Tests: NdjsonParserTests deleted; timestamp/critical-path/max-lifetime
tests repointed to EventIngestProcessor; the two NDJSON pipeline tests
removed (coverage retained by direct-list siblings); catalog/no-store
InlineData rows updated.
Docs: architecture.md (endpoint tables, ingest flow diagram, solution
layout, test matrix), web SectionAgentInternals endpoint tables,
build.ps1 help text.
Azure: no config change needed — /api/agent is not in the client-cert
exclusion paths (agent/ingest was always cert-enforced) and the
/api/bootstrap prefix exclusion remains for the surviving bootstrap
routes (verified via ARM on autopilotmonitor-api).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- docs/architecture.md Diagnostics section rewritten: both destinations
(CustomerSas default + Hosted opt-in), new download routing on shape,
cascade-delete step §5b with SAS-delete-permission gating, and the V2
remote-config retry + observability events.
- Submodule .claude/commands bumped to 17a6f5b for the dual-destination
fetch-diag-zip.sh refactor + skill error-handling note updates.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds opt-in resolution of Intune Platform Script + Remediation display names in
session timelines via a tenant-side appRoleAssignment grant. No app manifest
change, no impact on customers who don't enable it. Architecture is reusable for
future optional Graph permissions: catalog -> detector -> resolver -> endpoint.
Backend
- GraphFeatureDetector parses the `roles` claim from the SP access token and
caches the (token, roles) bundle per tenant. GetSnapshotAsync surfaces a
transient flag so admin-UI callers can render "unknown" instead of a
misleading "0 of 0 granted" when token acquire times out.
- GraphTokenService now accepts a CancellationToken so optional-feature callers
can apply tight budgets instead of inheriting the 5+15+30 s retry chain.
- ScriptDisplayNameResolver: cache-first, list-pull-on-cold-tenant, per-ID
fallback. Throttle-aware abort: skip per-ID for kinds whose full-pull hit a
transient failure; break the per-ID loop on the first HttpRequestException.
Total budget 5 s, with a tight 1.5 s sub-budget on the token acquire.
- Azure Table `ScriptNameCache` + direct-TableClient fail-loud repository
(mirrors TableOffboardingAuditRepository pattern). Daily
ScriptNameCacheCleanupFunction: 7 d positive / 1 h NotFound TTL.
- POST /api/tenants/{id}/scripts/display-names (body-carried refs to dodge URL
length limits; backend cap 200 refs/request).
- GET /api/tenants/{id}/graph-permissions/status (4 s budget; includes app
ClientId so the UI can render an exact grant command).
- POST /api/tenants/{id}/graph-permissions/refresh (invalidates token cache).
- All three routes registered in EndpointAccessPolicyCatalog (MemberRead for the
resolve, TenantAdminOrGA for the admin-UI surfaces).
Web
- `useScriptDisplayNames` hook with stable refset fingerprint (no spurious
refetches on parent re-renders); clears stale map on tenant/refset change so
a session switch never leaks names across tenants.
- Chunked POST loop: 200-ref batches with partial-success semantics. Preserves
earlier chunks when later ones fail (non-2xx OR transport throw); only
TokenExpiredError propagates so auth flow can react.
- ScriptExecutions renders " . {DisplayName}" alongside existing label/short-id,
keyed by Kind + policyId so Platform/Remediation under the same id never
collide.
- New Settings section "Optional Graph capabilities": status table with granted
/ not granted / unknown, copy-paste PS one-liner with ClientId baked in,
refresh button gated on response.ok.
Customer-facing
- scripts/CustomerSetup/Grant-AutopilotMonitorAddOn.ps1: idempotent
grant / verify / revoke script. -Features ScriptDisplayNames (high level) or
-Permissions ... (admin-UI generated). Pure ASCII for PS 5.1 compat.
- docs/customer-graph-add-on-permissions.md: end-to-end customer guide
covering prerequisites, the appRoleAssignment pattern, and revocation.
Tests
- 1718 backend xUnit / 364 web vitest. Highlights: JWT roles parsing edge
cases; full-pull-429 suppresses per-ID; per-ID-429 breaks loop; transient
detection via GetSnapshotAsync; chunked partial success (HTTP failure + thrown
transport); TokenExpiredError propagation; EndpointPolicyCatalogCompleteness
covers the three new routes by reflection.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Oliver Kieselbach