Advisor runs, screen settings and mini-program consent

Observed work

The advisor moved toward a single cloud-function scheduling path. Manual and automatic runs share a last-run-based interval, with a ten-minute minimum. Pending proposals are kept per point: a changed action replaces the old one, while an unchanged action extends its validity. The separate cooldown and per-run success-count cap were removed from the selected design, with one action per point per run retained. Implementation and test reports cover the models, commands and UI; they do not establish complete production execution.

A saved, sufficiently configured advisor can now be run manually while disabled. Description, point selection, AI configuration and entitlement checks remain. The UI reports that a request is queued rather than already started. One direct cloud-function attempt timed out on Redis before claiming its run. A later screen showed a successful run, but the evidence does not connect that success to a verified dependency repair. A test-environment migration also hit a timeout, with its final retry result uncaptured.

History retrieval was narrowed after a run consumed a large input context. Sensor summaries begin at the last successful run, ignore failed runs and cover between ten minutes and six hours; a first run uses six hours. Sampling, statistics and raw readings remain available when needed, while controllable points use operation and alert records. A read-only production query exposed the wrong value field; the corrected query returned data, followed by reports of 185 advisor tests and later 1,145 backend tests passing. The reorganized code was reportedly merged and pushed. Server and cloud-function deployment was still in progress. A separate interpretation error treated the summary-window start as gateway recovery time; its proposed wording repair had no final result.

In the evening, laan reported that an advisor run claimed to have switched an output off while the output still appeared on. An investigation was requested. No diagnosis, repair or repeat check is available, so the account retains the discrepancy without assigning a cause.

Industrial-screen settings work reduced the steps around point groups: bulk selection, separate scrolling areas and fixed controls were added, and the layout-preview entry was removed. Incompatible saved layouts fall back to automatic layout. Display switches save immediately, block repeated input during saving and restore the last confirmed state on failure. Targeted groups of five, five and nine tests were reported passing, with some changes pushed. Shared sticky-page framing was still being implemented; project switching, access validation and scene-image saving lacked complete final checks.

The screen application received a revised industrial-screen-shaped icon and resource/package checks. A tool report records Cloud Screen 0.1.20 being installed over 0.1.19 on the intended screen. Its App Info icon appeared correctly without clipping. The launcher entry was not separately accepted.

The Windows serial investigation identified connection before device selection as the relevant user path. The local entry flow now requires device selection first. YenGear Toolbox 2.8.35 also received icon changes and passing test, build, package and checksum reports. Uploading that version and connecting it to real Windows hardware were not verified; the earlier timeout is not recorded as resolved.

Mini-program privacy and login work settled on anonymous access to Home only, using a six-hour snapshot. Devices, projects and deeper routes still require login. Explicit consent precedes WeChat login, with agreement links on a separate line. Version 2.1.6 was reported uploaded without being submitted for review. The requested target was then corrected to 2.1.8; rebuilding, re-uploading and the last consent-layout check had no captured completion.

An online gateway check reported receipt of named point configuration and continued telemetry. UMS and App deployment had not accompanied that check. Alarm action declarations, tests and simulation continued moving into module packages, with local edits remaining. The report does not establish physical control and readback for every model or acceptance of the complete paired release.