Normalize whitespace in required wiring snippets so gofmt alignment changes
do not fail the model-only handoff contract. Preserve the same required
identifiers and forbidden adapter checks.
Refs #1782
Return persisted planning acceptance or refusal inside the investigation turn.
Keep model judgment separate from action authority and preserve accepted action
identity across provider failures. Enforce actor/request idempotency atomically
and retain complete approval and independent verification context.
Preserve unknown disk evidence, stream whitespace and historical resolution
timestamps. Keep conversation scrolling inside its own panel. Record real-model,
disposable-lab and browser qualification with explicit population limits.
Refs #1782
Keep the shared tool classification contract while removing assertions that
require the retired session state machine and inferred recovery behavior.
Refs #1782
Proxmox planning and dispatch now require a unique credential-admitted typed
runner with durable receipts. Development authentication preserves explicit
bearer identity so runner activation keeps its tenant and credential scope.
Preserve observer and receiver timestamps from their separate clocks instead
of rejecting or rewriting valid evidence. Keep completed execution separate
from stale or inconclusive verification, and label independent observations
accurately in action reviews.
Verified with targeted race suites, action-review tests and frontend build,
plus a real Assistant start plan and approved VM110 start/stop with independent
Proxmox confirmation. Final action reviews passed Playwright at 1440, 900 and
390 pixels, including retained completed, rejected and expired history.
Expose confined, identity-bound filesystem observations through the shared
resource pipeline so investigations can distinguish an exhausted container
mount from unrelated host capacity. Keep unavailable measurements explicit.
Isolate alert-history reads from durable writes and reuse one chronological
fold across polling. Catch up through bounded durable event IDs so simultaneous
readers do not replay every retained snapshot. Retain expired actions when
investigation outcomes move back to needs attention, and keep attached
Assistant context focused.
Record live storage diagnosis, healthy and dependency controls, approved and
rejected Docker outcomes, source-bound browser proof and exact test limits.
Missing-access continuity, VM dispatch completion and remaining Assistant
orchestration defects stay open in the redesign plan.
Refresh durable investigation lifecycle from authoritative actions while
preserving completed evidence. Keep resolved history reviewable and label
recorded plan facts separately from action outcomes. Follow all resource
pages during qualification and record live approval, rejection and storage
semantic-review results. Integrate current main and preserve its alert
ordering correction.
Live funded qualification found hidden tool results and misleading action
submission outcomes. Share the result-bearing transcript across stored chat
and product history, render the retained evidence, and distinguish captured
proposals from broker acceptance. Keep review usable while Patrol is paused.
Record Gemini route pricing and exact qualification limits. Integrate current
main and repeat browser proof for the incoming login flow. Approved/rejected
recovery remains unqualified without the development command agent.
The fleet sampler and coordinator had no production alert trigger and could
repeat cached values as fresh incident evidence. Preserve saved recordings
through explicit read-only lookups, propagate read failures and report the
former live incident count as unmeasured. Keep historical status and duration
units explicit without rewriting archived observations.
Integrate main's alert dispatch wording and startup replay qualification.
Canonical incident listing and real-model outcome qualification remain open.
Missing block I/O and container image sizes could become false evidence
for diagnosis. Preserve per-direction counter presence and measured zero
through collection, resource conversion and browser rendering. Separate
new observed history from ambiguous retained disk series without deleting
old rows or changing public metric names.
Keep partial host rates distinct and persist a newly enabled Disk I/O
column across the first preference reload.
Route labels classify ASCII digits and hexadecimal UUID bytes. Scan those
bytes directly while preserving numeric precedence, Unicode names and
invalid UTF-8 handling. This reduces the shared normaliser overhead exposed
by paired landing benchmarks without changing the benchmark gate.
Refs #1928
Contract-Neutral: ASCII route-label optimization preserves label values, identifier precedence, Unicode and invalid UTF-8 behavior, and agent lifecycle authority.
Integrate the latest alert, delivery and action-result changes with the
Patrol evidence conversation. Replace the conflicted browser receipt
with current source-bound qualification and fix shared warning-card
wrapping exposed by the intermediate-width check.
Real diagnostic and autonomous action outcome qualification stays open.
Remove contextless evaluator and assessment passes, signal-count budgets,
and post-finding prompt replacement. Keep evidence tools available until
explicit run limits and retain incomplete assessments and provider errors
alongside accepted decisions. Failed file reads now preserve error status
through the model, telemetry and saved Assistant history.
Full chat, AI and tools packages, focused Patrol API and race tests pass.
Real read-only and scripted browser checks preserve failed reads and linked
uncertainty. Real-model and verified action outcome qualification remain open.
Keep canonical disk risk, source freshness and retained history intact when
Assistant and Patrol gather evidence. Proposal acceptance validates an action
contract and must not rewrite uncertain conclusions as established root cause.
Preserve complete subscription tool batches without exposing routing envelopes
as answers. Keep wide answer tables readable and keyboard-scrollable on mobile.
Optimize retained tier reconciliation without discarding gaps or newer samples.
Record failed real-model diagnoses and outstanding autonomous qualification
separately from passing data-path and interface checks.
Replacement identity alone could confirm an unhealthy or restarting container when no independent observer was available. Check agent-reported running state and health, retaining inconclusive truth for unknown health and preserving intentionally stopped updates. Seven regression cases cover these distinctions; the four unsafe cases failed before the repair. Execution and compensation history are unchanged.
Change-source: pulse-maintainer
Independent update verification must match the replacement's reported state and reject unhealthy running observations, not just match container identity. Preserve intentionally stopped replacements and keep missing agent readback inconclusive. Reproduced stopped, restarting and unhealthy false confirmations before the fix; 19 focused API tests and three lifecycle replay tests pass. This does not establish the root cause of issue #1891.
Change-source: pulse-maintainer
Real homelab investigations lost metrics behind mismatched resource IDs,
queried only post-restart memory, and filtered every physical disk out.
Resolve canonical metrics targets, use retained history, preserve CPU
topology and chronological samples, and separate disk operation and format.
Keep unattached resource shortcuts blocked while allowing named resources
to be resolved through query search. Report incomplete readiness honestly,
including saved evaluations, without granting Patrol suitability.
Scoped local sessions correctly lose privileged security-status fields, but Settings relied on the configured admin username and hid owner controls. Expose the validated current principal without broadening capabilities, and use it for organisation identity.
Include the API and dependent subsystem contracts, recognised payload and settings-shell tests, and a source-bound desktop/narrow browser receipt in this commit. Fresh matrices each pass six scenarios with one expected disabled-feature skip. Admission/reconnect and interruption cleanup remain separate unfinished qualification.
Change-source: pulse-maintainer
Some identity providers publish unsupported Ed448 or X448 keys alongside supported signing keys. go-oidc v3.20 rejects the whole set, blocking otherwise valid logins; v3.21 skips only those unsupported keys while retaining normal token verification, and the Pulse service regression test pins that behaviour.
Contract-Neutral: dependency-only OIDC interoperability fix; no product or runtime contract changed
Change-source: pulse-maintainer
Core E2E shards aggregate many isolated browser sessions behind one Docker
bridge address, which can exhaust the production-sized general API limiter and
hide later assertions behind HTTP 429 responses. Raise only the explicitly
enabled development harness limit and keep malformed, unsafe, or production
overrides fail-closed.
Align the offline Proxmox journey with the reviewed native-button disclosure
ownership so adjacent links are checked against the actual disclosure control.
Change-source: pulse-maintainer
Contract-Neutral: Development-only validation isolation and regression alignment; production API limits and product behavior are unchanged.
The customers most likely to lapse are the ones who stopped opening Pulse,
so the in-app "This week" card cannot reach them. Report schedules gain a
kind: the default resources kind is the existing PDF or CSV performance
report, and patrol_digest emails the same weekly "what Patrol did for you"
rollup the Patrol page shows, in plain language, for the whole workspace.
It reuses the existing scheduler, cadence, recipients, tenant email config,
and Pro advanced-reporting entitlement rather than adding a second
scheduler. Digest schedules are weekly and email-only; the server fixes
scope, format, and attachments, and a run with no AI service, no Patrol, or
no email destination records a last_error that names the missing piece
instead of sending an empty week. Nothing is written to disk.
The digest assembly moves from the HTTP handler into
AISettingsHandler.BuildPatrolDigest so the schedule runner and the
endpoint share one code path. docs/MSP.md documents the kind for
providers; the Settings > Reporting "Report type" selector follows in a
separate change once it has a Pro-licensed browser pass.
The 2026-09-01 Patrol assessment could not answer four questions from
telemetry: what share of Patrol installs run local versus cloud models,
what Patrol costs an install per month (issue 1789 estimates ~104k input
tokens per full run at ~5.5 runs a day, above the Pro fee on non-Flash
tiers), how the 240 of 255 fleet investigations that produced no plan
actually ended, and which effective Patrol mode an install runs.
Schema v17 adds closed buckets only. ai_provider_class classifies the
Patrol model route (none, local, cloud_byok, cloud_subscription,
hosted_quickstart, unknown) with a syntactic private-host check on custom
OpenAI-compatible endpoints; it never resolves DNS or carries a provider,
model, endpoint, or account. pulse_intelligence_patrol_autonomy_level is
the effective level after licence and Autopilot gating. The two token
fields bucket the existing local usage ledger's Patrol events, so exact
counts and prices stay on the install. Thirteen investigation outcome
counters partition the findings already counted as investigated, one
bucket per finding, adding no finding, resource, or session identity.
The strings are never omitted, so an empty value can only mean a pre-v17
sender; the receiver stores those as unknown. PRIVACY.md, its shipped
mirror, the Settings preview interface, and the security-privacy and
api-contracts subsystem notes carry the disclosure, and the coverage gap
is registered in status.json.
The agent-lifecycle and storage-recovery contracts name internal/api under
their extension points, so both record that this telemetry is adjacent
adoption analytics and never agent or storage state. The stable E2E tier
pins the public schema version, so the disclosure spec moves to 17, and
the Settings preview interface change carries a browser receipt from a
scratch build of this tree at 1280x800 and 390x844. status.json also drops
an identical duplicate of the ai-provider-guided-setup coverage-gap record
that two merges each appended, which the private governance audit rejects.
The bucket for a legacy config still pointing at the retired Pulse-hosted
route is named hosted_legacy: the shipped privacy document discloses the
vocabulary verbatim and the frontend copy contract keeps that document
free of the retired hosted quickstart wording, so the telemetry package
now pins the vocabulary at the source.
Nothing in Pulse showed a paying customer what Patrol had done for them:
about 164 runs a month per install, findings raised, investigations and
fixes, and none of it summarised. GET /api/ai/patrol/digest rolls the last
N days (default 7, max 30) up from records Pulse already keeps: run
history, the findings store, Patrol-origin action audits, and the usage
cost store. It adds no telemetry and persists nothing. The payload
reports when the bounded run history no longer covers the window and when
model pricing is unknown, rather than quietly under-counting.
This is the first slice of the "Patrol weekly digest" named bet in the
pulse-pro demand ledger; the in-app "This week" card follows once its
browser pass is recorded. docs/PATROL_WEEKLY_DIGEST.md holds the design
note and the honest limits of each line.
status.json registers the patrol-value-visibility coverage gap, the
candidate lane, and its work claim. It also drops the second, identical
copy of the ai-provider-guided-setup coverage gap that landed with #1853;
the duplicate id fails the status audit on main for every pull request.
Forty-two percent of paying installs never get Patrol running, and the
support and issue evidence says the wall is model choice and cost: a
prospect asked which models Pulse recommends before trying, a Pro customer
picked Flash-Lite to save money and Patrol then could not file verdicts,
and a mispriced Opus row tripped the 20 USD budget so the circuit breaker
disabled Patrol with nothing but a log line (#1789).
Answer both questions at the point of choice. The Patrol and shared
default model pickers pin guided models in a "Suggested for Patrol"
section with recommended / suggested / caution markers and a one-line
reason: the Ollama preflight blessing is the only recommended entry,
Gemini Flash-Lite the only caution, and cloud starting points are labelled
price-driven and unqualified until this install's own readiness pass
upgrades them to "Verified on this install". A server-computed cost
preview sits under the model choice: monthly estimate from Pulse's price
table, the schedule, and the install's own median full run once three
priced runs exist (a measured 104,528-in / 4,491-out run otherwise), with
the assumption stated and a token explained once, plus 30-day spend
against budget. Each schedule preset is priced in the Schedule select.
When a per-token model is picked while the schedule is still the 6-hour
default, the cost model proposes the slowest preset that keeps scheduled
runs under half the budget (20 USD reference when none is set) and says
what the slower check costs in detection delay; a schedule the install
already chose is never changed.
A used-up budget is now a spending decision, not a provider fault: the
budget refusal is a typed sentinel with the figures, classified as the
budget_exhausted cause, excluded from circuit-breaker accounting, and
promoted into the Patrol block state, so the Patrol page shows "Patrol
paused" with the spend and limit and a "Raise the cost budget" action
instead of "Check Patrol model".
The Docker analog of the #1753 estate was still broken: two live
machines reusing one short hostname and one pasted unified install
token collapsed into a single flip-flopping DockerHost record, because
the hostname+token identity fallback adopts a record whose machine ID
disagrees with the report's. That fold is deliberate for recreated
containers (whose /etc/machine-id regenerates), so it cannot simply be
guarded by machine-ID inequality - the discriminator is a revisit: a
recreated container transitions to its new machine ID exactly once,
while two live machines alternate. Removing the collapsed record then
revoked the shared token unconditionally, rejecting every surviving
module - host reports included, since a unified install shares one
credential - with 401 "Unauthorized access attempt".
Consult the identity flap tracker before the hostname fallbacks adopt a
machine-ID-disagreeing record: a report whose machine ID returns to a
value already seen behind that identity is a second live machine and is
not folded. The machine whose identifiers minted the record reclaims it,
so the first site keeps its record and history, and the other site falls
through to the token binding check, converging on the documented "Each
Docker / Podman module must use a unique API token" rejection instead of
silently overwriting the record every cycle. RemoveDockerHost now skips
token revocation while any host or Docker record still authenticates
with the credential, mirroring the host-agent removal guard.
Regression coverage: an end-to-end router test walking the two-site
shared-token estate (host + Docker reports, alternating cycles, removal)
asserting the first site's identity stays stable, the second site gets
the unique-token guidance, and the shared token survives removal; a
router test proving removal of a machine's Docker record keeps the
unified token its host record still uses; and a state-layer test pinning
the reclaim/no-flip-flop convergence. Recreated-container adoption and
the existing token-uniqueness rejections keep their tests unchanged.
Two standalone Proxmox sites that reuse one short node name and one
pasted install token still folded into a single host record: the
token+hostname binding and the continuity store's hostname+token
fallback both resolve identity without consulting the machine ID, so
the second site's reports adopted the first site's host ID and the
record flip-flopped between machines. Removing that collapsed record
(the agent uninstaller does this on reinstall) then revoked the shared
token because the still-used guard could only see one record, leaving
every surviving agent rejected with 401 "Unauthorized access attempt"
on /api/agents/agent/report - the new symptom reported on #1753 after
the v6.4.2 retest.
Qualify host token binding slots with the reporting machine's identity
so two machines behind one token+hostname hold separate bindings,
refuse to adopt a bound or continuity identity whose recorded machine
ID disagrees with the report's, and rebuild bindings machine-qualified
after token reloads. Reports without a machine ID keep the legacy
binding shape and same-machine reinstalls keep their stable IDs.
Regression coverage: a state-layer test proving two shared-token pve01
machines keep distinct stable identities, continuity-match tests for
the machine guard, and an end-to-end router test walking the reporter's
estate - both install patterns, alternating reports, uninstall and
reinstall - asserting report auth never breaks and no token is revoked
while its sibling site still uses it.
The group-writable executable check wrote its fixture through os.WriteFile,
which filters the mode through the process umask, so under the runner's 022
umask the group-write bit never reached disk and validateTrustedExecutable
correctly trusted the file. Chmod the fixture to the asserted mode.
The Proxmox guest executor tests stamped their after observations one second
past the test-start clock, but actionStartedAt is stamped inside
ExecuteAction after handler setup, so a loaded runner overran the margin and
the independent observation was discarded as pre-action. Widen the
observation offset to a minute.
Both failures broke build-and-test on main (run 33454838531). Reproduced the
hostagent failure on Linux under umask 022 and verified both packages green
after the fix.
Three consecutive main runs were red because landed features outran their
guardrails while change detection kept the affected shards out of their own
runs.
- Resync shipped copies of API.md, CODE_SIGNING_POLICY.md, and
PRODUCTION_SECURITY.md from repo docs.
- Track the schema v16 telemetry disclosure (workload-history adoption
counts) in the settings architecture guardrail.
- Track route-scoped TrueNAS hydration queries and the conditional Kind
column class in the platform overview layout guardrails.
- Track the contained command-authority copy (legacy combined command
profile) in the installer and setup guide tests.
- Match the sentence-split delivery retention copy in AlertDeliveryLogCard.
- Provide selectedAgentUrl, insecureMode, and customCaPath in the
InfrastructureWorkspace operations mock for the safe-collector commands.
- Include the resource list facets block in the two API contract snapshots,
normalizing the direct-marshal fixture the way every handler does.
- Require the fail-closed installer checksum wording in the
AGENT_SUBSTRATE doc contract.
rcourtman