v0.4.37

multica-ai/multicav0.4.37Aug 31, 2026by github-actions[bot]

AI Summary

Focuses on WeCom integration improvements, performance optimizations, and bug fixes for chat, dashboard, and desktop components.

Key Highlights

WeCom reply routing to the replica holding the bot's socket
Performance improvements with HTTP server timeouts and skill-file batching
Extensive i18n localization for status, priority, squad, and editor labels
Desktop daemon recovery for unexpected stops

New Features

Huawei Cloud CodeArts support
Native iPad support for mobile

Full Release Notes

## Changelog
* 7c0e750452599ef5ead020c8c787b32c512f2cd4 MUL-6569 feat(wecom): give every undelivered reply a reason an operator can count (#7428)
* 1dd6b9ecdbaa991bfd51b48ef5c269056045d547 MUL-6570 feat(wecom): route a reply to the replica that holds the bot's socket (#7429)
* 64ec7f54163d918d5d7fd4dcae857f241b7842d0 MUL-6737: fix(timeouts): close the two gaps the 2h inactivity budget left open (#7699)
* 07bc8bc6bc480b4c024efe07dd5f6f06422ed82b MUL-6739 fix(chat): keep the viewport at the live end while replies stream (#7612)
* d540eb6fe7f6f2f255520d6bd3a453bb3e13282b MUL-6783: perf(server): set read-header and idle timeouts on main HTTP server (#7683)
* be93feb0631ae6200f280a538d2a4d92dcbfbf27 MUL-6792: perf(server): batch skill-file loads in LoadAgentSkills (#7689)
* be8fb7cebf00352816819beef8a616bb669977b0 MUL-6795 fix(dashboard): contain analytics leaderboard horizontal overflow (#7700)
* 9a57d689b4462d50f5eb165492d39c43f8998bed MUL-6796 fix(issues): keep mobile activity status compact (#7703)
* 9556613b211814640e4e8ed23e999ea0589f54be MUL-6835 fix(i18n): localize status and priority values (#7754)
* 915803ed3bb7acf40d900dfe9419f7b2dd77c533 MUL-6838 fix(i18n): localize squad and editor labels (#7758)
* 6231962e9840e2f3bd67c9331050e3e4d69c2aea MUL-6840 fix(pi): preserve sessions across workdir changes (#7760)
* c1e1f11e21cc2ee7dca0ac506a21cb64cca87af4 MUL-6850 fix(i18n): localize onboarding status label and lint the attribute/toast copy path (#7776)
* 2ebe0ae327df8e9150564ca907492360caaffe57 MUL-6862: fix(codex): widen thread setup handshake budget (#7790)
* a5c331bce35bfc9225556b3d4385667ba3abcf93 MUL-6868 fix(migrations): repair partial VCS link schema (#7796)
* f93855470ff36871eebd3f24690bb2e032c7fc81 MUL-6874 fix(desktop): recover unexpectedly stopped daemon (#7808)
* db86c75e129e00fac7fba0c180c71ea1d6cfbfac MUL-6878: avoid per-row issue status resolution (#7814)
* 0a7992c20ca81ed2adbc0b9bde2480766616d8e1 MUL-6879 fix(chat): don't paint the chat where the newest message was predicted (#7816)
* 66f612008735a629c7353f80d3e39c9af0538ee1 MUL-6879 fix(chat): stop the bottom-stick fighting Virtuoso's height estimate (#7811)
* 3e55ce4db033c79108f775ae4d32a2060d2b300b MUL-6882: remove unused active database samplers (#7817)
* eddb0dc2ad1ea18da9a49ea86d43b988dc897bba MUL-6884 feat(metrics): attribute agent_runtime lookups by product source (#7822)
* 7c1d826a98513fa5bc27df0a638329073078d57f chore(metrics): remove database-sampled metrics (#7819)
* 0742b50428c6752db91f601b7c2bed38446d5a25 docs(changelog): add v0.4.37 release entry (2026-08-31) (MUL-6890) (#7823)
* fe592621882055b0204e0d89c4db1a187c154652 feat(agent): add Huawei Cloud CodeArts support (#6985)
* 9c8c7f039c0ada380f51c946959a9b6cc22dc164 feat(mobile): enable native iPad support (#7793)
* 79559ebb92c48746d716db30a85acdc8c3cef8ec fix(ci): unblock the release by fixing two flaky main failures (MUL-6890) (#7828)
* 30bb3747fe3aeb3ede893fdb92db142ef94ba392 fix(i18n): align zh-Hans copy with glossary (#7759)
* bfe32f5e39eab65e75043fb402efad2dd63b40c7 fix(runtime): wake daemon for local skill requests (#7713)
* 15280617bc264e367ca9c5e5e5cefdb0988246b7 fix(settings): stop claiming workspace delete removes all data (#7771)
* 8d7e6f6a0a6fca901e07bcb5ee3d2dbd43393dcb perf(server): resolve only the requested skill bundles (#7810)
* 88e9170dc142d4e7ee03d2842bc676276a5995bf refactor(redact): drop home-directory masking from Text (#7772)