Agent Breaking Changes_

← Qwen Code

Qwen Code Release v0.26.0

2026-10-21 · official notes

Summary

• <!-- qwen-release-notes:v1 -->
• See the complete change list below.
• Breaking Changes

Full notes

<!-- qwen-release-notes:v1 -->

## Highlights

_See the complete change list below._

## Breaking Changes

No known breaking changes.

## Complete Change List

### Features

- feat(daemon): support session-targeted Skill and Extension refresh ([#14050](https://github.com/QwenLM/qwen-code/pull/14050)) by @callmeYe
- feat(runtime-broker): Session-create warm for Workspace-isolated bound Sessions ([#14599](https://github.com/QwenLM/qwen-code/pull/14599)) by @doudouOUC
- feat(managed-agent): ship a configuration template in the reference distribution ([#14631](https://github.com/QwenLM/qwen-code/pull/14631)) by @wenshao
- feat(serve): run private CSI file turns through Hosted Harness ([#14487](https://github.com/QwenLM/qwen-code/pull/14487)) by @doudouOUC
- feat(managed-agent): support local and SSH workspace file references ([#14639](https://github.com/QwenLM/qwen-code/pull/14639)) by @doudouOUC
- Adds a read-only --check preflight that prints PASS, WARN, FAIL, or SKIP for managed-agent deployment checks and exits nonzero on failure. ([#14632](https://github.com/QwenLM/qwen-code/pull/14632)) by @wenshao
- Lets a successor Hosted owner finish private CSI file turns that already have durable receipts, preserving tool responses and avoiding repeated file operations. ([#14523](https://github.com/QwenLM/qwen-code/pull/14523)) by @doudouOUC
- Managed Agent submissions now accept up to 8 MiB of attachment input, preserve large pending requests across refresh, and let users discard unrecoverable drafts. ([#14628](https://github.com/QwenLM/qwen-code/pull/14628)) by @doudouOUC
- Hosted Turn recovery now asks the Runtime Broker for execution evidence and treats unprovable outcomes as unknown instead of transient faults. ([#14669](https://github.com/QwenLM/qwen-code/pull/14669)) by @wenshao
- Adds systemd units and setup guidance for single-host managed-agent deployments so Runtime workers stay running when Broker restarts ([#14633](https://github.com/QwenLM/qwen-code/pull/14633)) by @wenshao
- Parks managed-agent turns with unknown Runtime outcomes for later rechecks instead of repeatedly retrying cancellation, keeping them active and avoiding false FAILED results ([#14681](https://github.com/QwenLM/qwen-code/pull/14681)) by @wenshao
- feat(artifacts): record files the agent creates or edits ([#14292](https://github.com/QwenLM/qwen-code/pull/14292)) by @zjgzx1988
- Adds a managed-agent credential bootstrap that generates required deployment tokens and writes protected configuration to simplify secure setup. ([#14634](https://github.com/QwenLM/qwen-code/pull/14634)) by @wenshao
- Completes private CSI cold recovery so settled file work can resume from saved results without repeating file effects or extra model requests. ([#14695](https://github.com/QwenLM/qwen-code/pull/14695)) by @doudouOUC
- Adds an optional Generated UI extension that renders model-generated HTML/CSS cards with bounded local JavaScript, letting users adjust inputs without another model turn. ([#14338](https://github.com/QwenLM/qwen-code/pull/14338)) by @qqqys
- feat(workflow): authorize daemon ancestor workflows ([#14646](https://github.com/QwenLM/qwen-code/pull/14646)) by @qqqys
- feat(web-shell): show per-model TTFT and output speed in Token Usage ([#14728](https://github.com/QwenLM/qwen-code/pull/14728)) by @qqqys
- feat(runtime): add original CSI application retirement ([#14698](https://github.com/QwenLM/qwen-code/pull/14698)) by @doudouOUC
- feat(telemetry): Record reasoning effort on LLM spans ([#14718](https://github.com/QwenLM/qwen-code/pull/14718)) by @zonemeen
- feat(serve): add local daemon discovery and control commands ([#14739](https://github.com/QwenLM/qwen-code/pull/14739)) by @wenshao
- feat(web-shell): Polish advanced table controls and copying ([#14707](https://github.com/QwenLM/qwen-code/pull/14707)) by @ytahdn
- feat(core): account tool result sizes from producer to injection ([#14545](https://github.com/QwenLM/qwen-code/pull/14545)) by @yiliang114
- feat(web-shell): add trajectory follow and pause controls (PR7B) ([#14723](https://github.com/QwenLM/qwen-code/pull/14723)) by @qqqys
- feat(web-shell): delete selected message groups from session history ([#14297](https://github.com/QwenLM/qwen-code/pull/14297)) by @ytahdn
- feat(core): honor preview budgets and make truncated previews recoverable ([#14546](https://github.com/QwenLM/qwen-code/pull/14546)) by @yiliang114
- feat(web-shell): Preview collapsed Home content on hover ([#14826](https://github.com/QwenLM/qwen-code/pull/14826)) by @ytahdn
- feat(web-shell): preview PDFs with the native browser viewer ([#14825](https://github.com/QwenLM/qwen-code/pull/14825)) by @ytahdn
- feat(web-shell): improve token usage hierarchy ([#14648](https://github.com/QwenLM/qwen-code/pull/14648)) by @ytahdn
- feat(mods): support tool middleware in interactive terminals ([#14565](https://github.com/QwenLM/qwen-code/pull/14565)) by @qqqys
- feat(web-shell): show background shell task runtime details ([#14831](https://github.com/QwenLM/qwen-code/pull/14831)) by @ytahdn
- feat(core): report why loop protection stopped a turn ([#14858](https://github.com/QwenLM/qwen-code/pull/14858)) by @qqqys
- feat(serve): add qwen daemon start ([#14869](https://github.com/QwenLM/qwen-code/pull/14869)) by @wenshao
- feat(channels): add WeCom application callback support ([#13976](https://github.com/QwenLM/qwen-code/pull/13976)) by @callmeYe
- feat(serve): add daemon status detail ([#14902](https://github.com/QwenLM/qwen-code/pull/14902)) by @wenshao
- feat(cli): expose experimental Mods in settings ([#14889](https://github.com/QwenLM/qwen-code/pull/14889)) by @qqqys
- feat(web-shell): preview DOCX files across file entry points ([#14868](https://github.com/QwenLM/qwen-code/pull/14868)) by @ytahdn
- feat(web-shell): refresh model catalogs from the model selector ([#14685](https://github.com/QwenLM/qwen-code/pull/14685)) by @callmeYe
- feat(serve): add daemon start --wait ([#14914](https://github.com/QwenLM/qwen-code/pull/14914)) by @wenshao
- feat(managed-agent): expose session MCP catalog in WebShell ([#14916](https://github.com/QwenLM/qwen-code/pull/14916)) by @doudouOUC
- feat(runtime): add explicit idle resident Workspace retirement ([#14765](https://github.com/QwenLM/qwen-code/pull/14765)) by @doudouOUC
- feat(web-shell): consolidate trajectory chrome into a summary bar and toolbar ([#14913](https://github.com/QwenLM/qwen-code/pull/14913)) by @qqqys
- feat(managed-agent): Support approval changes during a turn ([#14920](https://github.com/QwenLM/qwen-code/pull/14920)) by @doudouOUC
- feat(managed-agent): simplify existing local project onboarding ([#14970](https://github.com/QwenLM/qwen-code/pull/14970)) by @doudouOUC
- feat(generated-ui): bundle Generated UI behind an experimental flag ([#14888](https://github.com/QwenLM/qwen-code/pull/14888)) by @qqqys
- feat(web-shell): show model refresh as an icon beside search ([#14983](https://github.com/QwenLM/qwen-code/pull/14983)) by @callmeYe
- feat(managed-agent): import enabled skills at session creation ([#14930](https://github.com/QwenLM/qwen-code/pull/14930)) by @doudouOUC
- feat(serve): add daemon restart ([#14929](https://github.com/QwenLM/qwen-code/pull/14929)) by @wenshao
- feat(cua): streamline computer use and port native reliability fixes ([#14846](https://github.com/QwenLM/qwen-code/pull/14846)) by @LaZzyMan
- feat(web-shell): support managed session rename ([#14987](https://github.com/QwenLM/qwen-code/pull/14987)) by @doudouOUC
- feat(serve): add persisted workspace artifact library ([#14867](https://github.com/QwenLM/qwen-code/pull/14867)) by @BZ-D
- feat(workflow): add per-turn human-origin workflow hints ([#14876](https://github.com/QwenLM/qwen-code/pull/14876)) by @qqqys
- feat(web-shell): open trajectory record details as a drill-in or side panel ([#14984](https://github.com/QwenLM/qwen-code/pull/14984)) by @qqqys
- feat(core): add a real PTY backend for the Bun runtime ([#14046](https://github.com/QwenLM/qwen-code/pull/14046)) by @chiga0
- feat(managed-agent): allow concurrent sessions on shared workspace storage ([#15026](https://github.com/QwenLM/qwen-code/pull/15026)) by @doudouOUC
- feat(managed-agent): derive Session Store wiring on a single host ([#15005](https://github.com/QwenLM/qwen-code/pull/15005)) by @wenshao
- feat(serve): make paired Web Shell devices listable and revocable ([#15000](https://github.com/QwenLM/qwen-code/pull/15000)) by @yiliang114
- feat(core): add model.loopProtection and remind instead of stopping loops by default ([#15016](https://github.com/QwenLM/qwen-code/pull/15016)) by @qqqys
- feat(cli): paste clipboard text with Ctrl+V on Linux ([#14172](https://github.com/QwenLM/qwen-code/pull/14172)) by @qwen-code-dev-bot
- feat(web-shell): unify daemon and managed workspace navigation ([#15061](https://github.com/QwenLM/qwen-code/pull/15061)) by @doudouOUC

### Bug Fixes

- fix(managed-agent): bound initial publication guard waits ([#14563](https://github.com/QwenLM/qwen-code/pull/14563)) by @yiliang114
- test(cli): retry ENOTEMPTY in hosted harness teardowns ([#14636](https://github.com/QwenLM/qwen-code/pull/14636)) by @ZedingZhang
- Automatically recovers public Sessions with committed, unconsumed inputs after managed-agent writer loss so pending wakes are replayed instead of dropped. ([#14603](https://github.com/QwenLM/qwen-code/pull/14603)) by @doudouOUC
- Session creation and worktree reset via the daemon HTTP API now use the daemon's session budget deadline, preventing in-progress initialization from timing out too early. ([#14588](https://github.com/QwenLM/qwen-code/pull/14588)) by @yiliang114
- On POSIX, Hosted stdio MCP servers now run in their own process group with supervised shutdown, improving cleanup; Windows behavior is unchanged. ([#14640](https://github.com/QwenLM/qwen-code/pull/14640)) by @wenshao
- Hosted Session journals now recover after a lapsed writer lease, applying committed entries once and retrying missing ones instead of failing. ([#14570](https://github.com/QwenLM/qwen-code/pull/14570)) by @wenshao
- Hosted PostToolUse, PostToolUseFailure, and PostToolBatch hooks now report the tool input left by PreToolUse, excluding refused or invalid replacements. ([#14650](https://github.com/QwenLM/qwen-code/pull/14650)) by @qqqys
- fix(ci): repair Java gates and report failures on PRs ([#14522](https://github.com/QwenLM/qwen-code/pull/14522)) by @yiliang114
- Hosted sessions now continue recovered child-result and monitor notification turns after a crash instead of waiting for new input. ([#14679](https://github.com/QwenLM/qwen-code/pull/14679)) by @doudouOUC
- Raises Playwright wait limits for web-shell smoke tests from 10 to 20 seconds to reduce false failures on busy shared CI hosts ([#14643](https://github.com/QwenLM/qwen-code/pull/14643)) by @yiliang114
- Fixes managed-agent tool cards so expanding a settled ReadFile or Shell card loads saved arguments, output, and errors ([#14604](https://github.com/QwenLM/qwen-code/pull/14604)) by @doudouOUC
- Preserves keyboard focus in the web-shell trajectory panel when recent execution views are removed or evicted ([#14641](https://github.com/QwenLM/qwen-code/pull/14641)) by @qqqys
- fix(cli): reconcile managed recovery failure reasons ([#14567](https://github.com/QwenLM/qwen-code/pull/14567)) by @yiliang114
- fix(hooks): reap a Hook command unit when its worker dies (#14502) ([#14675](https://github.com/QwenLM/qwen-code/pull/14675)) by @wenshao
- fix(config): let a Managed host restore its own session ([#14688](https://github.com/QwenLM/qwen-code/pull/14688)) by @wenshao
- Widens interactive E2E wait budgets for saved workflow slash-command completion replies to 120 seconds to reduce flaky CI stalls. ([#14202](https://github.com/QwenLM/qwen-code/pull/14202)) by @qwen-code-dev-bot
- Prevents a pending user cancellation from being recorded as a failed turn when recovery is not ready, allowing it to remain cancelling or settle as cancelled. ([#14694](https://github.com/QwenLM/qwen-code/pull/14694)) by @wenshao
- Bounds managed-agent publication placement waits to prevent unlimited blocking and long lock timeouts during dispatch, completion, claim, and verification. ([#14542](https://github.com/QwenLM/qwen-code/pull/14542)) by @yiliang114
- Managed Agent image attachments now accept PNG, JPEG, GIF, WebP, and the image/jpg alias, matching Qwen Code image handling. ([#14704](https://github.com/QwenLM/qwen-code/pull/14704)) by @doudouOUC
- Stabilizes the FG12b released-store fault gate so it reliably verifies owner-release resend behavior without depending on CI load. ([#14711](https://github.com/QwenLM/qwen-code/pull/14711)) by @qwen-code-dev-bot
- Fixes CI so Java failure summaries can be posted as pull request comments by granting the comment writer job pull-request write permission. ([#14717](https://github.com/QwenLM/qwen-code/pull/14717)) by @yiliang114
- fix(managed-agent): reject an incomplete Workspace tool path in --check ([#14635](https://github.com/QwenLM/qwen-code/pull/14635)) by @wenshao
- fix(cli): hold pre-runtime /acp WebSocket upgrades instead of falling through to a 406 (#13995) ([#13997](https://github.com/QwenLM/qwen-code/pull/13997)) by @qwen-code-dev-bot
- fix: align DeepSeek reasoning controls across official and MaaS routes ([#14689](https://github.com/QwenLM/qwen-code/pull/14689)) by @qqqys
- fix(web-shell): surface the fix witness in the review artifact viewer ([#14725](https://github.com/QwenLM/qwen-code/pull/14725)) by @yiliang114
- fix(managed-agent): align fault gates with unsettled-append recovery ([#14716](https://github.com/QwenLM/qwen-code/pull/14716)) by @yiliang114
- test(managed-agent): widen the hosted recovery and hook-cancel gate windows (#14729) ([#14741](https://github.com/QwenLM/qwen-code/pull/14741)) by @qwen-code-dev-bot
- fix(web-shell): preserve slash highlighting in user messages ([#14171](https://github.com/QwenLM/qwen-code/pull/14171)) by @Lilian0122
- fix(managed-agent): prove a zombie-only group stopped ([#14738](https://github.com/QwenLM/qwen-code/pull/14738)) by @wenshao
- test(managed-agent): drive the FG12b releasing gate until the fault fires (#14782) ([#14799](https://github.com/QwenLM/qwen-code/pull/14799)) by @qwen-code-dev-bot
- test(managed-agent): pin the FG13c worker-kill reaper contract ([#14802](https://github.com/QwenLM/qwen-code/pull/14802)) by @qwen-code-dev-bot
- test(managed-agent): give the failed-spawn test a realistic deadline (#14779) ([#14783](https://github.com/QwenLM/qwen-code/pull/14783)) by @qwen-code-dev-bot
- fix(core): Compact managed activation and inbox journals ([#14760](https://github.com/QwenLM/qwen-code/pull/14760)) by @doudouOUC
- fix(managed-agent): Reconcile parked Turn cancels from durable results ([#14763](https://github.com/QwenLM/qwen-code/pull/14763)) by @doudouOUC
- fix(ci): repair main's Prettier step and deflake the hook-cancel fence test (#14581) ([#14590](https://github.com/QwenLM/qwen-code/pull/14590)) by @qwen-code-dev-bot
- fix(ci): wait 90 minutes for the sdk-java host lock (#14787) ([#14801](https://github.com/QwenLM/qwen-code/pull/14801)) by @qwen-code-dev-bot
- perf(core): Index managed Session history reads ([#14767](https://github.com/QwenLM/qwen-code/pull/14767)) by @doudouOUC
- fix(managed-agent): park a cancel over an unobserved Turn (D7-2) ([#14700](https://github.com/QwenLM/qwen-code/pull/14700)) by @wenshao
- fix(ci): include updatedInput hook integration tests in PR checks ([#14747](https://github.com/QwenLM/qwen-code/pull/14747)) by @yiliang114
- fix(ci): allow fault gates to outlive the Surefire fork timeout ([#14745](https://github.com/QwenLM/qwen-code/pull/14745)) by @yiliang114
- fix(cli): keep one-shot system reminders out of the user's own message ([#14740](https://github.com/QwenLM/qwen-code/pull/14740)) by @yiliang114
- fix(ci): raise the build heap cap from 4096 to 6144 MB ([#14822](https://github.com/QwenLM/qwen-code/pull/14822)) by @yiliang114
- fix(acp): avoid splitting surrogate pairs when truncating turn text ([#14744](https://github.com/QwenLM/qwen-code/pull/14744)) by @yiliang114
- fix(managed-agent): Preserve original owners for owed Hosted cleanup ([#14764](https://github.com/QwenLM/qwen-code/pull/14764)) by @doudouOUC
- fix(core): keep a producer-declared truncation bit when raw size is unknown ([#14830](https://github.com/QwenLM/qwen-code/pull/14830)) by @yiliang114
- fix(ci): cap the prepare build heap and share the self-hosted Node preflight ([#14827](https://github.com/QwenLM/qwen-code/pull/14827)) by @wenshao
- fix(ci): skip failure reports for superseded run attempts (#14768) ([#14771](https://github.com/QwenLM/qwen-code/pull/14771)) by @qwen-code-dev-bot
- fix(managed-agent): answer self-healing Harness conflicts as retryable 503 ([#14645](https://github.com/QwenLM/qwen-code/pull/14645)) by @wenshao
- fix(web-shell): avoid flashing the credential form during boot token verification ([#14770](https://github.com/QwenLM/qwen-code/pull/14770)) by @BigDaddy1337
- fix(core): stop double-truncating web_fetch and keep the full text recoverable ([#14547](https://github.com/QwenLM/qwen-code/pull/14547)) by @yiliang114
- fix(managed-agent): keep the CSI retirement key mode-independent on MariaDB (#14832) ([#14853](https://github.com/QwenLM/qwen-code/pull/14853)) by @yiliang114
- fix(hosted): bind execution to the final hook input ([#14708](https://github.com/QwenLM/qwen-code/pull/14708)) by @qqqys
- test(managed-agent): stabilize the FG12b intent and cwd hosted gates ([#14730](https://github.com/QwenLM/qwen-code/pull/14730)) by @qwen-code-dev-bot
- fix(managed-agent): treat hosted_turn_active as a lifecycle wait (#14492) ([#14806](https://github.com/QwenLM/qwen-code/pull/14806)) by @qwen-code-dev-bot
- fix(ci): retry transient pnpm installs in the SDK Java workflow (#14845) ([#14851](https://github.com/QwenLM/qwen-code/pull/14851)) by @qwen-code-dev-bot
- fix(channels): preserve Markdown structure when splitting messages ([#14043](https://github.com/QwenLM/qwen-code/pull/14043)) by @callmeYe
- fix(memory): report the real cause and drop set when MEMORY.md exceeds its budget ([#14841](https://github.com/QwenLM/qwen-code/pull/14841)) by @yiliang114
- fix(ci): skip Managed Agent reviews after the PR closes (#14774) ([#14777](https://github.com/QwenLM/qwen-code/pull/14777)) by @qwen-code-dev-bot
- fix(managed-agent): bound settled runtime dispatch entries ([#14657](https://github.com/QwenLM/qwen-code/pull/14657)) by @wenshao
- fix(ci): skip prepare's duplicate build in the daemon E2E install (#14861) ([#14873](https://github.com/QwenLM/qwen-code/pull/14873)) by @qwen-code-dev-bot
- fix(cli): keep workspace skill status aligned after session load ([#14840](https://github.com/QwenLM/qwen-code/pull/14840)) by @callmeYe
- test(managed-agent): Measure the post-commit gate as per-round growth (#14197) ([#14207](https://github.com/QwenLM/qwen-code/pull/14207)) by @qwen-code-dev-bot
- fix(managed-agent): refresh schema upgrade expectations for V12/V13 ([#14893](https://github.com/QwenLM/qwen-code/pull/14893)) by @qwen-code-dev-bot
- fix(managed-agent): bound and harden session resource storage ([#14656](https://github.com/QwenLM/qwen-code/pull/14656)) by @wenshao
- fix(managed-agent): stop activation renewal after persistent failure ([#14658](https://github.com/QwenLM/qwen-code/pull/14658)) by @wenshao
- test(cli): give resident model recovery waits the slow-host budget (#14922) ([#14927](https://github.com/QwenLM/qwen-code/pull/14927)) by @qwen-code-dev-bot
- fix(ci): hold the MariaDB lane's sdk-java lock once per job (#14896) ([#14903](https://github.com/QwenLM/qwen-code/pull/14903)) by @qwen-code-dev-bot
- fix(ci): hold chosen RocketMQ ports until the JVMs bind them ([#14860](https://github.com/QwenLM/qwen-code/pull/14860)) by @yiliang114
- fix(release): publish the generated CLI with npm so vendored ripgrep keeps its exec bit ([#14907](https://github.