Qwen Code Desktop v0.28.0 is a Alibaba release published 2026-11-17 with 110 listed changes, flagged for security fixes. Summary and full notes below, extracted from the official notes.
One free email: every breaking change, retirement and price change across 23 coding agents. No spam, unsubscribe anytime.
Subscribe free← All Qwen Code releases and breaking changes
<!-- Release notes generated using configuration in .github/release.yml at 895dfca8ff125ac0e16372edacd4653322c02147 --> ## What's Changed * feat(web-shell): add sidebar workspace filter by @callmeYe in https://github.com/QwenLM/qwen-code/pull/17244 * feat(managed-agent): add the DingTalk adapter on the managed channel path by @wenshao in https://github.com/QwenLM/qwen-code/pull/17136 * fix(managed-agent): settle stale public-child recovery as terminal instead of retrying forever by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/17273 * fix(sdk): route Skill mutations through native REST by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17111 * feat(acp): route live and read-only Agent Host sessions through the shared tool batch by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/17218 * fix(sdk): keep the capabilities deadline through body reads by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17103 * fix(sdk): release settled WebSocket probe deadlines by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17107 * fix(cli): preserve surfaced task selection across reordering by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17126 * fix(cli): ignore unchanged JSON todo snapshots by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17122 * fix(web-shell): preserve withheld memory content as read-only by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17109 * fix(review): propagate explicit run deadlines to the child by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17118 * fix(cli): preserve the auth dialog exit confirmation window by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17125 * fix(cli): show disabled multiselect rows by selection state by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17124 * fix(core): retain buffered shell output after process exit by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17112 * feat(core): let the loop switches be forced off, not only on by @2SinceJuly in https://github.com/QwenLM/qwen-code/pull/17048 * fix(web-shell): clear Goals multi-workspace post-merge review items by @wenshao in https://github.com/QwenLM/qwen-code/pull/17291 * chore(deps): update vulnerable dependencies by @LubabaKhalid in https://github.com/QwenLM/qwen-code/pull/13169 * test(cli): cover the shared batch host inputs on cron and notification turns by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/17330 * docs(design): add the agent loop's capability ledger and rollout rules by @qqqys in https://github.com/QwenLM/qwen-code/pull/17285 * feat(web-shell): make the local files bridge easier to discover and use by @wenshao in https://github.com/QwenLM/qwen-code/pull/17198 * docs(serve): document the two ignored watch knobs and calibrate the bound-delivery designs by @wenshao in https://github.com/QwenLM/qwen-code/pull/17309 * feat: make each surface's resolved agent-loop path visible by @wenshao in https://github.com/QwenLM/qwen-code/pull/17256 * fix(core): recheck transcript file type on the opened fd (TOCTOU hardening) by @wenshao in https://github.com/QwenLM/qwen-code/pull/17301 * fix(web-shell): finalize voice dictation before sending a message by @wenshao in https://github.com/QwenLM/qwen-code/pull/17331 * fix(acp): skip unusable transcript refreshes during active loads by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17106 * fix(cli): hide root-restricted workspace settings by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17121 * test(web-shell): stabilize global turn navigation scroll-follow E2E (#17252) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17259 * fix(sdk): refresh replay capability after transport reconnects by @Afloat16 in https://github.com/QwenLM/qwen-code/pull/17110 * chore(deps): clear remaining CVE audit findings (diff DoS, uuid bounds check) by @now-ing in https://github.com/QwenLM/qwen-code/pull/10985 * test(web-shell): stabilize the workflow-mode policy read test (#17348) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17355 * fix(cli): record cancelled tool results on the headless shared loop by @qqqys in https://github.com/QwenLM/qwen-code/pull/17353 * feat(managed-agent): enable external workspace execution by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/15684 * test(core): deflake the unawaited-timer code mode test by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17345 * feat(verify-pr): require a boundary declaration and own the watch state by @wenshao in https://github.com/QwenLM/qwen-code/pull/17185 * test(integration): settle the ACP parity debug log before stopping the agent by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17365 * fix(cli): preserve hosted turn history after interruption by @water-in-stone in https://github.com/QwenLM/qwen-code/pull/15959 * feat(channels): digest bound-PR comment backlog after a downtime gap by @wenshao in https://github.com/QwenLM/qwen-code/pull/17199 * test(ci): stabilize the macOS nightly CI legs by @wenshao in https://github.com/QwenLM/qwen-code/pull/17219 * feat(core): run headless sessions and the Ink TUI on the shared agent loop by default by @qqqys in https://github.com/QwenLM/qwen-code/pull/17362 * fix(channels): land the managed-channel review follow-ups (#17173) by @wenshao in https://github.com/QwenLM/qwen-code/pull/17351 * feat(serve): localize the daemon PR-watch prompt and pin the recorded follow-ups by @wenshao in https://github.com/QwenLM/qwen-code/pull/17352 * docs(design): record speculation PR 6b decision to keep its own tool step by @wenshao in https://github.com/QwenLM/qwen-code/pull/17324 * ci(release): retry the Docker integration test suites once by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/16862 * fix(cli): preserve full commands in execution confirmations by @chiga0 in https://github.com/QwenLM/qwen-code/pull/17361 * test(core): cover speculation's partial-batch boundary hand-back on both arms by @wenshao in https://github.com/QwenLM/qwen-code/pull/17399 * docs(design): close the owed claims #17352 landed and the stale role-guidance status by @wenshao in https://github.com/QwenLM/qwen-code/pull/17395 * fix(channels): evict refused DingTalk webhooks and surface admission diagnostics by @wenshao in https://github.com/QwenLM/qwen-code/pull/17396 * docs(users): add Web Shell voice, Live, and image generation guides by @wenshao in https://github.com/QwenLM/qwen-code/pull/17347 * docs(design): record the landings the channel design statuses still hide by @wenshao in https://github.com/QwenLM/qwen-code/pull/17409 * test(cli): wait for the applied skills selection before navigating by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17413 * test(channels): pin the managed DingTalk lifecycle, settings, and export surfaces by @wenshao in https://github.com/QwenLM/qwen-code/pull/17417 * feat(managed-agent): publish the managed /loop skill variant by profile by @wenshao in https://github.com/QwenLM/qwen-code/pull/17272 * fix(cli): narrow /effort to the tiers the active model supports by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/17425 * fix(verify-pr): close the context trust refusal's root-tmp and hard-link bypasses by @wenshao in https://github.com/QwenLM/qwen-code/pull/17375 * feat(web-shell): Complete the Russian Web Shell UI catalog by @callmeYe in https://github.com/QwenLM/qwen-code/pull/17414 * test(channels): pin the managed DingTalk stream wiring and webhook send budget by @wenshao in https://github.com/QwenLM/qwen-code/pull/17426 * docs(design): bring the rollout-switch design up to date with main by @qqqys in https://github.com/QwenLM/qwen-code/pull/17410 * refactor(managed-agent): archive Kubernetes runtime and resident CSI backend by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17415 * fix(cli): let a rerun join go on with the credential its first run saved by @wenshao in https://github.com/QwenLM/qwen-code/pull/16831 * docs(design): reconcile the unified-loop end state with the keep-its-own decisions by @wenshao in https://github.com/QwenLM/qwen-code/pull/17416 * feat(core): retain the latest complete operation during compaction by @callmeYe in https://github.com/QwenLM/qwen-code/pull/17141 * feat(web-shell): add active highlight styling to selected reasoning e… by @pranavshankar1221 in https://github.com/QwenLM/qwen-code/pull/16622 * test(generated-ui): read comma-joined amounts and drop an unobservable check by @qqqys in https://github.com/QwenLM/qwen-code/pull/17430 * feat(core): run subagents on the shared agent loop by default by @qqqys in https://github.com/QwenLM/qwen-code/pull/17420 * feat(web-shell): show per-turn hook statistics by @callmeYe in https://github.com/QwenLM/qwen-code/pull/17327 * test(core): characterize leader delivery and terminal state per host class by @tonydzi in https://github.com/QwenLM/qwen-code/pull/14855 * feat(acp): let sessions a program drives take accepted cross-session messages by @qqqys in https://github.com/QwenLM/qwen-code/pull/17436 * feat(mods): let Mods rewrite expanded thinking in the terminals by @qqqys in https://github.com/QwenLM/qwen-code/pull/17443 * fix(channels): guard every managed DingTalk persist by its side-effect position by @wenshao in https://github.com/QwenLM/qwen-code/pull/17432 * feat(core): slim edit output and add authorized stale recovery by @callmeYe in https://github.com/QwenLM/qwen-code/pull/17255 * fix(managed-agent): Wait for workspace preparation before turn admission by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17457 * fix(serve): accept hook description and statusMessage during extension preparation by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17439 * fix(managed-agent): read the content-only approval selection at public-child version 1 by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/17428 * feat(web-shell): surface a session-artifact header entry before the replay on cold open by @wenshao in https://github.com/QwenLM/qwen-code/pull/17397 * fix(managed-agent): resolve external execution actor through child session lineage by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/17412 * test(managed-agent): add a real model-turn E2E for the /loop skill by @wenshao in https://github.com/QwenLM/qwen-code/pull/17433 * feat(generated-ui): store card view state in the session transcript by @qqqys in https://github.com/QwenLM/qwen-code/pull/17448 * feat(web-shell): make Goal state and session navigation readable at a glance by @wenshao in https://github.com/QwenLM/qwen-code/pull/17349 * feat(web-shell): resolve dirty-tree checkout refusals in the branch picker by @wenshao in https://github.com/QwenLM/qwen-code/pull/17364 * test(managed-runtime): pin the deferred acquire against close by @2SinceJuly in https://github.com/QwenLM/qwen-code/pull/16492 * test(serve): pin description and statusMessage acceptance on managed hooks by @MohammedAlkindi in https://github.com/QwenLM/qwen-code/pull/17393 * feat(serve): run a create_sub_session sub-session in another workspace by @qqqys in https://github.com/QwenLM/qwen-code/pull/17454 * fix(managed-agent): resolve shared workspace owners for extension control by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17480 * fix(managed-agent): recover macOS workspaces after reboot by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17466 * fix(managed-agent): recover timed-out hosted journal commits by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17459 * fix(web-shell): classify the directory-collision checkout refusal and close post-merge gaps by @wenshao in https://github.com/QwenLM/qwen-code/pull/17476 * fix(channels): retire the DingTalk SDK heartbeat interval on socket close by @wenshao in https://github.com/QwenLM/qwen-code/pull/17462 * test(web-shell): deflake large source file smoke test by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17373 * fix(cli): avoid watching unrelated settings entries by @chiga0 in https://github.com/QwenLM/qwen-code/pull/17475 * test(managed-agent): restore the round-3 /loop E2E fixes and wire the scenario into CI by @wenshao in https://github.com/QwenLM/qwen-code/pull/17470 * feat(web-shell): add delivery-unit change overview for sessions by @wenshao in https://github.com/QwenLM/qwen-code/pull/17378 * ci(review): opt the review's own commands into the execution sandbox by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/17282 * refactor(managed-agent): unify coordinator dispatch deduplication and lease renewal by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17508 * refactor(managed-agent): replace regex DDL catalog parsing with a tokenizing parser by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17509 * feat(channels): add Aone Code channel plugin with polling event-stream inbound by @wenshao in https://github.com/QwenLM/qwen-code/pull/17442 * docs(design): propose a unified chat surface for daemon and managed sessions by @wenshao in https://github.com/QwenLM/qwen-code/pull/17460 * chore(managed-agent): deduplicate verification Failsafe profiles and land #17139 follow-ups by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17507 * feat(web-shell): Codex-style managed worktree sessions (default-on, detached, branch & handoff) by @wenshao in https://github.com/QwenLM/qwen-code/pull/17228 * feat(serve): let create_sub_session follow up a sub-session the caller created by @qqqys in https://github.com/QwenLM/qwen-code/pull/17484 * test(web-shell): restore the label-only adoption witness and reconcile the classifier's documented reach by @wenshao in https://github.com/QwenLM/qwen-code/pull/17503 * fix(channels): give each DingTalk reply chunk its own send budget by @wenshao in https://github.com/QwenLM/qwen-code/pull/17504 * feat(web-shell): add activity date session grouping by @callmeYe in https://github.com/QwenLM/qwen-code/pull/17447 * fix(deps): patch express-rate-limit window accounting by @chiga0 in https://github.com/QwenLM/qwen-code/pull/17525 * test(cli): bound managed terminal watchdogs and teardown by host budget by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17268 * fix(core): accept native ripgrep regex syntax in Grep by @dakjdakd in https://github.com/QwenLM/qwen-code/pull/14315 * feat: add qwen4-max support with native effort ladder, cache TTL, and image detail by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17513 * refactor(managed-agent): require canonical Broker URLs at the preflight and Harness entry by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/17510 * test(integration): settle the second turn before reloading the generated-UI session by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/17537 ## New Contributors * @LubabaKhalid made their first contribution in https://github.com/QwenLM/qwen-code/pull/13169 * @pranavshankar1221 made their first contribution in https://github.com/QwenLM/qwen-code/pull/16622 **Full Changelog**: https://github.com/QwenLM/qwen-code/compare/desktop-v0.27.4...desktop-v0.28.0