Qwen Code v0.27.1 is a Alibaba release published 2026-11-09 with 126 listed changes, flagged for breaking changes. 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
<!-- qwen-release-notes:v1 --> ## Highlights _See the complete change list below._ ## Breaking Changes No known breaking changes. ## Complete Change List ### Features - feat(core): name the active shell in the startup context ([#16491](https://github.com/QwenLM/qwen-code/pull/16491)) by @2SinceJuly - feat(qqbot): add opt-in forwardBotMessages for bot-authored messages ([#15904](https://github.com/QwenLM/qwen-code/pull/15904)) by @Eric-GoodBoy-Tech - feat(agents): make default background execution configurable ([#15178](https://github.com/QwenLM/qwen-code/pull/15178)) by @ZedingZhang - feat(cli): enable LSP through a user-scoped setting and env pair ([#14613](https://github.com/QwenLM/qwen-code/pull/14613)) by @BigDaddy1337 - feat(channels): give bound sessions role guidance without operator templates ([#16576](https://github.com/QwenLM/qwen-code/pull/16576)) by @wenshao - feat(ci): switch review runner day pool to a numeric count over hz runners ([#16544](https://github.com/QwenLM/qwen-code/pull/16544)) by @yiliang114 - feat(ci): support HTTP proxy in ephemeral runner containers ([#16592](https://github.com/QwenLM/qwen-code/pull/16592)) by @yiliang114 - feat(serve): add the daemon PR watch store foundation ([#16590](https://github.com/QwenLM/qwen-code/pull/16590)) by @wenshao - feat(acp-bridge): expose replay projection watermark ([#16538](https://github.com/QwenLM/qwen-code/pull/16538)) by @chiga0 - feat(web-shell): expose Channel delivery targets in Scheduled Tasks ([#16524](https://github.com/QwenLM/qwen-code/pull/16524)) by @jkxiao - feat(web-shell): add workspace collapse controls and drag sorting ([#16197](https://github.com/QwenLM/qwen-code/pull/16197)) by @ytahdn - feat(web-shell): label rendered artifact cards Preview and strengthen the details UI-language prompt ([#16587](https://github.com/QwenLM/qwen-code/pull/16587)) by @wenshao - feat(daemon): expose last user prompt time and session sorting ([#14654](https://github.com/QwenLM/qwen-code/pull/14654)) by @samuelhsin - feat(serve): observe bound PRs and detect state transitions behind an off switch ([#16611](https://github.com/QwenLM/qwen-code/pull/16611)) by @wenshao - feat(managed-agent): add Session resource and agent-trace diagnostics ([#16640](https://github.com/QwenLM/qwen-code/pull/16640)) by @doudouOUC - feat(ci): give the autofix address agent failed-check evidence ([#16659](https://github.com/QwenLM/qwen-code/pull/16659)) by @wenshao - feat(serve): drain bound-PR watch pending transitions behind an off switch ([#16656](https://github.com/QwenLM/qwen-code/pull/16656)) by @wenshao - feat(managed-agent): persist Session model statistics and request context ([#16641](https://github.com/QwenLM/qwen-code/pull/16641)) by @doudouOUC - feat(managed-agent): add archive and MCP self-service installation ([#16500](https://github.com/QwenLM/qwen-code/pull/16500)) by @doudouOUC - feat(channels): protect the GitHub mention lane from watch budget spend ([#16701](https://github.com/QwenLM/qwen-code/pull/16701)) by @wenshao - feat(managed-agent): persist Session execution diagnostics ([#16642](https://github.com/QwenLM/qwen-code/pull/16642)) by @doudouOUC - feat(channels): deliver bound-PR comments to every session that did not post them ([#16602](https://github.com/QwenLM/qwen-code/pull/16602)) by @wenshao - feat(web-shell): show scheduled task workspace when the picker is hidden ([#16630](https://github.com/QwenLM/qwen-code/pull/16630)) by @jkxiao - feat(mods): run the prompts a Mod submits in the OpenTUI terminal ([#16426](https://github.com/QwenLM/qwen-code/pull/16426)) by @qqqys - feat(web-shell): add sidebar multi-select with bulk pin and archive ([#16702](https://github.com/QwenLM/qwen-code/pull/16702)) by @deggs7 - feat(managed-agent): add opt-in personal memory backed by Mem0 ([#16488](https://github.com/QwenLM/qwen-code/pull/16488)) by @doudouOUC ### Bug Fixes - fix(core): share one gh runner and name an unusable view directory ([#16545](https://github.com/QwenLM/qwen-code/pull/16545)) by @wenshao - fix(telemetry): report OpenAI cache_write_tokens as cache creation ([#16454](https://github.com/QwenLM/qwen-code/pull/16454)) by @2SinceJuly - fix(ci): route external PR CI directly to disposable runners ([#16551](https://github.com/QwenLM/qwen-code/pull/16551)) by @yiliang114 - fix(channels): skip memory extraction and follow-up suggestions on channel turns ([#16565](https://github.com/QwenLM/qwen-code/pull/16565)) by @wenshao - fix(ci): report timed-out main CI runs instead of letting cancelled hide them ([#16280](https://github.com/QwenLM/qwen-code/pull/16280)) by @yiliang114 - fix(web-shell): refresh standalone sessions after allocation ([#15311](https://github.com/QwenLM/qwen-code/pull/15311)) by @ZedingZhang - fix(batch): report provider-specific output budgets in readiness checks ([#15181](https://github.com/QwenLM/qwen-code/pull/15181)) by @Jun516 - fix(core): share compression cache for equivalent model selectors ([#15318](https://github.com/QwenLM/qwen-code/pull/15318)) by @xxwtiancai - fix(desktop): restore saved web previews ([#14597](https://github.com/QwenLM/qwen-code/pull/14597)) by @zonemeen - fix(managed-agent): create a never-attached Session for its first automation verb (#14548) ([#14561](https://github.com/QwenLM/qwen-code/pull/14561)) by @xxwtiancai - fix(memory): report why a background memory agent stopped ([#13466](https://github.com/QwenLM/qwen-code/pull/13466)) by @everyoneexe - fix(web-shell): do not persist rejected daemon URL credentials ([#12285](https://github.com/QwenLM/qwen-code/pull/12285)) by @lorenzozanee - fix(core): base web terminal backpressure on pending input ([#14313](https://github.com/QwenLM/qwen-code/pull/14313)) by @dakjdakd - fix(cli): emit headless results before draining memory ([#13885](https://github.com/QwenLM/qwen-code/pull/13885)) by @xxwtiancai - fix(lsp): Attribute diagnostics to explicit extension owners ([#14117](https://github.com/QwenLM/qwen-code/pull/14117)) by @loglapa - fix(cli): settle Live call stops after failure ([#13835](https://github.com/QwenLM/qwen-code/pull/13835)) by @GoldArowana - fix(cli): remove dead REPLACE mergeStrategy declarations for modelProviders and providerProtocol ([#12613](https://github.com/QwenLM/qwen-code/pull/12613)) by @liwenjie200543 - fix(ci): run the failure patrol on ECS and time its sandbox image pull separately ([#16462](https://github.com/QwenLM/qwen-code/pull/16462)) by @wenshao - fix(managed-agent): let --check read the configuration the server reads ([#16364](https://github.com/QwenLM/qwen-code/pull/16364)) by @wenshao - fix(ci): keep sdk-java resolver bookkeeping and in-house jars out of the host cache (#15281) ([#15286](https://github.com/QwenLM/qwen-code/pull/15286)) by @qwen-code-dev-bot - fix(web-shell): narrow sidebarless live-state polling to displayed workspaces ([#16550](https://github.com/QwenLM/qwen-code/pull/16550)) by @yiliang114 - fix(cli): warn about Node 24/25 Wasm memory spikes without disabling AST checks ([#15321](https://github.com/QwenLM/qwen-code/pull/15321)) by @Willam2004 - fix(extension): persist CDP reconnect backoff across service worker restarts ([#14022](https://github.com/QwenLM/qwen-code/pull/14022)) by @he-yufeng - fix(web-shell): isolate mixed-direction Markdown text ([#14035](https://github.com/QwenLM/qwen-code/pull/14035)) by @hahahahahayesyeseys - fix(review): refuse the findings artifact by name and print the expected schema ([#14824](https://github.com/QwenLM/qwen-code/pull/14824)) by @ZedingZhang - fix(sdk-typescript): clamp an overflowing session-spawn deadline to the timer ceiling ([#14676](https://github.com/QwenLM/qwen-code/pull/14676)) by @Bug-killer-prog - fix(web-shell): route standalone reloads by session source type ([#12384](https://github.com/QwenLM/qwen-code/pull/12384)) by @cactuser-Lu - fix(ci): harden the already-published desktop probe ([#12184](https://github.com/QwenLM/qwen-code/pull/12184)) by @lorenzozanee - fix(web-shell): remove productScope gating on managed provider capabilities ([#16554](https://github.com/QwenLM/qwen-code/pull/16554)) by @yiliang114 - fix(managed-agent): prove A3 rewritten command recovery without Hook rerun ([#16116](https://github.com/QwenLM/qwen-code/pull/16116)) by @wenshao - fix(channels): journal GitHub mentions added by editing an older comment ([#16588](https://github.com/QwenLM/qwen-code/pull/16588)) by @wenshao - fix(serve): refuse to validate a restored worktree whose lock is not held ([#16593](https://github.com/QwenLM/qwen-code/pull/16593)) by @wenshao - fix(ci): check out all of .github for owner assignment ([#16599](https://github.com/QwenLM/qwen-code/pull/16599)) by @yiliang114 - fix(ci): stop ephemeral runner slots from wedging and alert on stalls ([#16597](https://github.com/QwenLM/qwen-code/pull/16597)) by @yiliang114 - fix(cli): accept explicitly named Local Control addresses outside RFC 1918/3927 ([#14140](https://github.com/QwenLM/qwen-code/pull/14140)) by @pomelo-nwu - fix(web-shell): keep footer actions stable during inline rename ([#15182](https://github.com/QwenLM/qwen-code/pull/15182)) by @Jun516 - fix(web-shell): show file changes in standalone sessions ([#15070](https://github.com/QwenLM/qwen-code/pull/15070)) by @zonemeen - fix(web-shell): keep the first prompt running when switching sessions during creation ([#16582](https://github.com/QwenLM/qwen-code/pull/16582)) by @wenshao - fix(mcp): name the capped server timeout in App read timeout warnings ([#13312](https://github.com/QwenLM/qwen-code/pull/13312)) by @tonydzi - fix(channels): read GitHub comments of the cursor's own second once ([#16606](https://github.com/QwenLM/qwen-code/pull/16606)) by @wenshao - fix(web-shell): scope file changes to the session workspace ([#16102](https://github.com/QwenLM/qwen-code/pull/16102)) by @ytahdn - fix(managed-agent): clear the SpotBugs null dereference in preflight ([#16600](https://github.com/QwenLM/qwen-code/pull/16600)) by @wenshao - fix(ci): stop cross-job qwen-home pollution from killing PR review ([#16618](https://github.com/QwenLM/qwen-code/pull/16618)) by @yiliang114 - fix(web-shell): reload an e2e boot aborted by host network churn ([#16595](https://github.com/QwenLM/qwen-code/pull/16595)) by @wenshao - fix(ci): exponential backoff on rate-limited runner registration ([#16608](https://github.com/QwenLM/qwen-code/pull/16608)) by @yiliang114 - fix(web-shell): label proxied daemon failures instead of an empty 500 ([#16610](https://github.com/QwenLM/qwen-code/pull/16610)) by @yiliang114 - fix(ci): cover hz hosts in the fleet qwen update and pin both prefixes ([#16619](https://github.com/QwenLM/qwen-code/pull/16619)) by @yiliang114 - fix(web-shell): close managed host integration gaps left by the host mount entry ([#16200](https://github.com/QwenLM/qwen-code/pull/16200)) by @yiliang114 - fix(managed-agent): fail preflight rows this process cannot verify ([#16639](https://github.com/QwenLM/qwen-code/pull/16639)) by @wenshao - fix(channels): post the GitHub error comment only after the last attempt ([#16631](https://github.com/QwenLM/qwen-code/pull/16631)) by @wenshao - fix(ci): quarantine a qwen-home neither rm nor sudo can reclaim ([#16627](https://github.com/QwenLM/qwen-code/pull/16627)) by @wenshao - fix(channels): stop paying for a CI aggregate a failure already decided ([#16665](https://github.com/QwenLM/qwen-code/pull/16665)) by @wenshao - fix(web-shell): keep Open for managed artifacts the default panel cannot preview ([#16645](https://github.com/QwenLM/qwen-code/pull/16645)) by @wenshao - fix(serve): clean a worktree checkout after more than one reset ([#16643](https://github.com/QwenLM/qwen-code/pull/16643)) by @wenshao - fix(web-shell): harden the dev daemon probe and proxy error attribution ([#16663](https://github.com/QwenLM/qwen-code/pull/16663)) by @yiliang114 - fix(web-shell): keep transcript visible during session resync ([#16688](https://github.com/QwenLM/qwen-code/pull/16688)) by @chiga0 - fix(ci): self-heal ephemeral runner slots wedged by orphaned resource piles ([#16628](https://github.com/QwenLM/qwen-code/pull/16628)) by @yiliang114 - fix(ci): add bounded, idempotent retry to the SDK release branch push ([#16705](https://github.com/QwenLM/qwen-code/pull/16705)) by @yiliang114 - fix(web-shell): let hosts own the composer add menu ([#16715](https://github.com/QwenLM/qwen-code/pull/16715)) by @yiliang114 - fix(managed): Allow short browser tokens in Web gateway ([#16723](https://github.com/QwenLM/qwen-code/pull/16723)) by @doudouOUC - fix(stats): include time to first token in output speed ([#16716](https://github.com/QwenLM/qwen-code/pull/16716)) by @qqqys - fix(ci): only delete label-recovered registrations GitHub reports offline ([#16713](https://github.com/QwenLM/qwen-code/pull/16713)) by @yiliang114 - fix(ci): stop staling branches for eslint ignores they never excluded ([#16706](https://github.com/QwenLM/qwen-code/pull/16706)) by @wenshao - fix(core): stop the renewal timer before reacquireWriter drops the grant ([#16720](https://github.com/QwenLM/qwen-code/pull/16720)) by @wenshao - fix(subscription): restore model switching and align session recovery ([#16738](https://github.com/QwenLM/qwen-code/pull/16738)) by @liuxuely0626 ### Performance - perf(managed-agent): read each restore resource once on a cold Workspace load ([#16625](https://github.com/QwenLM/qwen-code/pull/16625)) by @wenshao ### Documentation - docs(providers): document Wally under OpenAI-compatible providers ([#16529](https://github.com/QwenLM/qwen-code/pull/16529)) by @sanchitmonga22 - docs(daemon): document session-scoped permission vote for channel auto-approve ([#16541](https://github.com/QwenLM/qwen-code/pull/16541)) by @yiliang114 - docs(tools): add You.com free MCP to the web search MCP services ([#14000](https://github.com/QwenLM/qwen-code/pull/14000)) by @brainsparker - docs(channels): document self-pushed head echoes and scope their suppression ([#16577](https://github.com/QwenLM/qwen-code/pull/16577)) by @wenshao - docs(design): propose idle parking for resident Hosted Sessions ([#16564](https://github.com/QwenLM/qwen-code/pull/16564)) by @wenshao - docs(memory): document external context_search MCP setup ([#16041](https://github.com/QwenLM/qwen-code/pull/16041)) by @vshulcz - docs(serve): state the worktree lock key rule and the restore's lock-mismatch 500 ([#16601](https://github.com/QwenLM/qwen-code/pull/16601)) by @wenshao ### Internal Changes - test(cli): stop a flaky A2A approval poll from failing the whole shard ([#16555](https://github.com/QwenLM/qwen-code/pull/16555)) by @wenshao - refactor(serve): share the prompt admission with in-process callers ([#16543](https://github.com/QwenLM/qwen-code/pull/16543)) by @wenshao - ci: raise the ephemeral runner slot cap to 64 per host ([#16591](https://github.com/QwenLM/qwen-code/pull/16591)) by @yiliang114 - test(managed-agent): make the preflight tests pin what they claim ([#16589](https://github.com/QwenLM/qwen-code/pull/16589)) by @wenshao - test(serve): drive the reconcile probe through the Broker's real rejection path ([#16568](https://github.com/QwenLM/qwen-code/pull/16568)) by @2SinceJuly - refactor(serve): move the restore's worktree reconciliation into a helper ([#16583](https://github.com/QwenLM/qwen-code/pull/16583)) by @wenshao - test(acp-bridge): pin background activity and scope spy cleanup ([#12243](https://github.com/QwenLM/qwen-code/pull/12243)) by @Oscar-Williams - test(cli): pin that the trust store is forwarded, not mounted ([#16401](https://github.com/QwenLM/qwen-code/pull/16401)) by @qwen-code-dev-bot - test(ci): pin the stall alert to one per stall episode ([#16650](https://github.com/QwenLM/qwen-code/pull/16650)) by @yiliang114 - refactor(web-shell): share session view reset state ([#15060](https://github.com/QwenLM/qwen-code/pull/15060)) by @qwen-code-dev-bot - test(ci): keep the visuals publisher and the failure reporter off the pool ([#16561](https://github.com/QwenLM/qwen-code/pull/16561)) by @wenshao - ci: route the web-shell visuals capture for fork PRs to the disposable pool ([#16617](https://github.com/QwenLM/qwen-code/pull/16617)) by @yiliang114 - test(ci): pin the stall alert re-arm after a recovery ([#16660](https://github.com/QwenLM/qwen-code/pull/16660)) by @yiliang114 - test(managed-agent): give the diagnostics surface probes a session body ([#16694](https://github.com/QwenLM/qwen-code/pull/16694)) by @wenshao - test(cli): give two serve-side CI waits real poll budgets ([#16699](https://github.com/QwenLM/qwen-code/pull/16699)) by @wenshao - test(serve): pin the worktree ownership lock key derivations together ([#16646](https://github.com/QwenLM/qwen-code/pull/16646)) by @wenshao - test(ci): isolate the checkout hook fixture from the release runner git config ([#16684](https://github.com/QwenLM/qwen-code/pull/16684)) by @qwen-code-dev-bot - test(web-shell): let a boot retry outlast a network-change abort storm ([#16717](https://github.com/QwenLM/qwen-code/pull/16717)) by @wenshao - test(channels): pin the watch budget recovery edge against repeat rows ([#16721](https://github.com/QwenLM/qwen-code/pull/16721)) by @wenshao - test(cli): cover the Ink TUI's Goal, notification, cron and teammate turns ([#16704](https://github.com/QwenLM/qwen-code/pull/16704)) by @qqqys - test(cli): derive the upload cleanup budget from the lane test timeout ([#16711](https://github.com/QwenLM/qwen-code/pull/16711)) by @wenshao - test: deflake the temp-name watcher and auto-open disarm suites ([#16697](https://github.com/QwenLM/qwen-code/pull/16697)) by @wenshao - test(core): let the heavy title log test outlast a contended host ([#16730](https://github.com/QwenLM/qwen-code/pull/16730)) by @wenshao ## New Contributors - @sanchitmonga22 made their first contribution in [#16529](https://github.com/QwenLM/qwen-code/pull/16529) - @everyoneexe made their first contribution in [#13466](https://github.com/QwenLM/qwen-code/pull/13466) - @brainsparker made their first contribution in [#14000](https://github.com/QwenLM/qwen-code/pull/14000) - @vshulcz made their first contribution in [#16041](https://github.com/QwenLM/qwen-code/pull/16041) **Full Changelog**: https://github.com/QwenLM/qwen-code/compare/v0.27.0...v0.27.1