Qwen Code cua-driver-rs-v0.7.3 预发布

cua-driver prebuilt binaries — relative-coordinate fork (vendored under `packages/cua-driver`). - **macOS**: codesigned + notarized universal binary + `QwenCuaDriver.app` - **Linux**: unsigned (x86_64 + arm64, glibc 2.31 floor) - **Windows**: unsigned (x86_64 + arm64) Enable relative coordinates: `CUA_DRIVER_RS_COORDINATE_SPACE=1` (default `0` = off; optional `CUA_DRIVER_RS_COORDINATE_SCALE=1000`). Enable textual MCP payload filtering for affected model API routes: `MCP_MODEL_PAYLOAD_FILTER=1` (default off). <!-- Release notes generated using configuration in .github/release.yml at cb483092561a3606dfd53447a26f3718380f0f59 --> ## What's Changed * feat(review): prove Step 4 (verify) and Step 5 (reverse audit) actually ran by @wenshao in https://github.com/QwenLM/qwen-code/pull/6965 * ci: quarantine cron-interactive from push E2E to nightly-only by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6986 * fix(cron): add deterministic test seam for cron-interactive E2E by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/6987 * feat(channels): support natural memory references by @qqqys in https://github.com/QwenLM/qwen-code/pull/6952 * chore: update default model to qwen3.7-max by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6978 * feat(daemon): add stateless generation SSE by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6947 * fix(test): widen model-response timeouts in SDK E2E tests for CI stability (#6979) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6985 * perf(review): scope Agent 7's build/test to the workspaces the diff changed by @wenshao in https://github.com/QwenLM/qwen-code/pull/6955 * test(web-shell): add mermaid, split-view + sidebar visual scenarios by @wenshao in https://github.com/QwenLM/qwen-code/pull/6964 * ci(serve): daemon A/B before/after preview on response-surface PRs by @wenshao in https://github.com/QwenLM/qwen-code/pull/6975 * fix(cli): Preserve channel startup failure details by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6950 * feat(daemon): Aggregate deep health across workspaces by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6961 * ci(web-shell): before/after visual previews, showing only changed views by @wenshao in https://github.com/QwenLM/qwen-code/pull/6963 * feat(serve): add workspace MCP management by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6954 * feat(web-shell): color-code each split pane by workspace by @wenshao in https://github.com/QwenLM/qwen-code/pull/6971 * fix(web-shell): filter sessions by source by @ytahdn in https://github.com/QwenLM/qwen-code/pull/6995 * fix(web-shell): land on the split's first pane when a shrink folds the split by @wenshao in https://github.com/QwenLM/qwen-code/pull/7000 * fix(headless): run concurrency-safe tool calls in parallel by @wenshao in https://github.com/QwenLM/qwen-code/pull/6993 * feat(review): fold the findings list into the verify/reverse-audit prompt by @wenshao in https://github.com/QwenLM/qwen-code/pull/6994 * fix(shell): handle command-specific exit codes by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7011 * docs: make local PR verification optional by @callmeYe in https://github.com/QwenLM/qwen-code/pull/7025 * feat(channels): tag daemon sessions with channel source by @xurik in https://github.com/QwenLM/qwen-code/pull/6991 * fix(core): preserve MCP OAuth challenges from HTTP handshakes by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7022 * revert: remove local PR verification gate by @callmeYe in https://github.com/QwenLM/qwen-code/pull/7031 * test(web-shell): add extensions-manager visual scenario by @wenshao in https://github.com/QwenLM/qwen-code/pull/6997 * feat(daemon): Trace cold first-session startup by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6907 * fix(serve): Harden multi-workspace ownership guards by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7005 * docs(serve): Close multi-workspace hardening gaps by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7019 * feat(web-shell): add archived session export by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6910 * fix: bound usage-only streams and abort on quit by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7038 * fix(acp): disambiguate model routes by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7028 * refactor(web-shell): drop redundant primary-workspace label by @wenshao in https://github.com/QwenLM/qwen-code/pull/7035 * test(web-shell): make visual-preview captures deterministic + add workspace-sidebar scenario by @wenshao in https://github.com/QwenLM/qwen-code/pull/7041 * fix(web-shell): batch transcript dispatch to avoid tab-return freeze by @wenshao in https://github.com/QwenLM/qwen-code/pull/7012 * fix(core): Require explicit approval to exit Plan mode by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6967 * ci(autofix): recover from generated-artifact CI gates and stop silent stalls by @wenshao in https://github.com/QwenLM/qwen-code/pull/6998 * fix(web-shell): use formatSettingCategory for fallback UI category by @wenshao in https://github.com/QwenLM/qwen-code/pull/7055 * feat(serve): add workspace session-info aggregate endpoint by @samuelhsin in https://github.com/QwenLM/qwen-code/pull/7077 * feat(web-shell): add skill management pages by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7018 * ci(autofix): run the schema gate from a trusted staged copy, not the branch tree by @wenshao in https://github.com/QwenLM/qwen-code/pull/7076 * fix(integration): add missing session_info to E2E capabilities baseline by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7091 * feat(cli): mouse text selection and copy in VP mode by @chiga0 in https://github.com/QwenLM/qwen-code/pull/6937 * test(web-shell): align workspace sidebar visual smoke by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7107 * fix(tui): pin MaxSizedBox rows and gate the pending backstop for show-more diff (#6809) by @azurecgx in https://github.com/QwenLM/qwen-code/pull/6957 * fix(integration): harden flaky interactive read-then-write test by @wenshao in https://github.com/QwenLM/qwen-code/pull/7105 * test(cli): isolate sandbox-relaunch tests from ambient QWEN_SANDBOX_IMAGE by @wenshao in https://github.com/QwenLM/qwen-code/pull/7093 * ci(autofix): treat Suggestion-level review findings as actionable per AGENTS.md by @wenshao in https://github.com/QwenLM/qwen-code/pull/7094 * feat(channels): stamp daemon sourceId with channel instance name on created sessions by @xurik in https://github.com/QwenLM/qwen-code/pull/7078 * feat(channels): confirm natural memory mutations by @qqqys in https://github.com/QwenLM/qwen-code/pull/7066 * feat(channels): save explicit multi-fact memory safely by @qqqys in https://github.com/QwenLM/qwen-code/pull/7092 * feat(serve): Complete legacy session workspace telemetry by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7003 * fix(core): retry empty tool-result continuations by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7039 * fix(core): align planning and response guidance by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7085 * feat(core): Enable artifact defaults and write reminders by @chiga0 in https://github.com/QwenLM/qwen-code/pull/7068 * feat(web-shell): suggest sending new-topic drafts in a new session by @carffuca in https://github.com/QwenLM/qwen-code/pull/7098 * fix(web-shell): render built-in tag icons by @callmeYe in https://github.com/QwenLM/qwen-code/pull/7024 * fix(core): force tool_choice in generateJson to prevent auto-mode classifier deadlock by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6929 * fix(ci): notify silent triage re-runs by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7079 * fix(cli): pop the kitty keyboard protocol after leaving the alternate screen by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7115 * fix(web-shell): optionally restart SSE after prompt admission by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7080 * test(serve): cover session-info regressions by @samuelhsin in https://github.com/QwenLM/qwen-code/pull/7083 * fix(core): accept subagents created after startup instead of rejecting on stale cache by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7112 * fix(cli): keep streaming code blocks intact when split across commits by @MikeWang0316tw in https://github.com/QwenLM/qwen-code/pull/7020 * fix(i18n): correct "extenison" typo in extension enable/disable scope help by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7057 * feat(tools): add formatDisplayPath() and wire grep/glob/ripGrep descriptions by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7050 * fix(core): make the per-turn tool-call cap adaptive by @wenshao in https://github.com/QwenLM/qwen-code/pull/7052 * feat(daemon): propagate prompt IDs to turn events by @makwingchi in https://github.com/QwenLM/qwen-code/pull/7082 * fix(vscode): preserve Electron Node mode for ACP launch by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7106 * docs(cua-driver): fix broken relative-coordinate design link by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7130 * fix(core): preserve complete skill descriptions by @callmeYe in https://github.com/QwenLM/qwen-code/pull/7032 * feat(web-shell): paginate restored session history by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7064 * fix(core): persist resolved subagent model in metadata by @ARE404 in https://github.com/QwenLM/qwen-code/pull/7104 * feat(web-shell): toggle the session sidebar with Cmd+B / Ctrl+B by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7135 * fix(web-shell): stop stacking duplicate copies when restoring prompt text by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7134 * feat(web-shell): add directory autocomplete to the Add Workspace dialog by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7125 * feat(cli): support same-turn message steering by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/7090 * fix(core): retry malformed repeated thinking tags by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7100 * fix(cli): correct misspelled handleUpdateRecieved -> handleUpdateReceived by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7132 * fix(cli): hide sticky task panel when agent is idle by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7062 * feat: support full-turn multimodal routing for image prompts by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7045 * fix(cli): keep the model override when a background notification drains by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7119 * fix(core): align prompt tool examples with schemas by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7088 * fix(cli): require a second Ctrl+C within 1s before a real SIGINT exits the TUI by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7129 * fix: keep quit and stream logging reliable by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7124 * fix(ui): keep SkillReviewDialog right border inside the dialog container by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7047 * fix(core): preserve existing work in system prompt by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7087 * docs(autofix): make bot PR comments bilingual with collapsed Chinese by @wenshao in https://github.com/QwenLM/qwen-code/pull/7137 * fix(mcp): terminate descendants after discovery timeout by @morluto in https://github.com/QwenLM/qwen-code/pull/6926 * feat(agents): support per-model sub-agent concurrency limits by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/6984 * fix(review): report what the transcripts prove; build the roster in one call by @wenshao in https://github.com/QwenLM/qwen-code/pull/7033 * fix(cli): establish extension store generation baseline on first read by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7072 * fix(core): remove ask_user_question from the Explore agent's toolset by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7133 * fix(web-shell): scope advanced table overlays by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7097 * feat(cli): add daemon Todo stop guard by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6945 * feat(cli): show active path in compact tool summaries by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7043 * fix(core): correct "supercedes" typo in plan-mode system prompt by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7058 * fix(ask-user-question): accept long headers and size chips to the container width by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/7063 * fix(core): canonicalize restrictive permission paths by @morluto in https://github.com/QwenLM/qwen-code/pull/6923 * fix(core): align system prompt with interaction mode by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7089 * fix(memory): resolve root symlinks in isAllowedMemoryPath before creation by @wenshao in https://github.com/QwenLM/qwen-code/pull/6842 * fix(core): add kimi-k3 token limits (1M context, 128K output) by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/7144 * fix(integration): use lenient assertion and harden poll in interactive file-system test by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7113 * feat(daemon): Profile ACP channel initialization by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7145 * fix(core): resolve a parameter expansion in command position to its command root by @wenshao in https://github.com/QwenLM/qwen-code/pull/7143 * fix(cli): make auto output language follow user input by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6953 * fix(core): respect enableManagedAutoMemory in memory availability by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/6941 * fix(cli): tighten VP-mode controls footprint and fix shell tool indicator overlap by @chiga0 in https://github.com/QwenLM/qwen-code/pull/6931 * fix(acp): resolve textual @ image paths by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7123 * feat(vscode): route logs to the Qwen Code Companion output channel by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7121 * feat(core): overhaul web_fetch — content fidelity, binary handling, security, and resilience by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/7146 * feat(channels): expose workspace-scoped observed contacts by @BenGuanRan in https://github.com/QwenLM/qwen-code/pull/7109 * test(cli): actually exercise the paste-workaround path in useKeypress by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7141 * fix: correct typos in comments, a tool description, and docs by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7131 * feat(web-shell): persist the split view across refresh, per tab by @wenshao in https://github.com/QwenLM/qwen-code/pull/7136 * fix: harden desktop MCP permission-request lifecycle so pending prompts never hang or leak by @mvanhorn in https://github.com/QwenLM/qwen-code/pull/7013 * fix(cli): correct misspelled migratedInMemorScopes -> migratedInMemoryScopes by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7140 * feat(web-shell): add a workspace Goals page, and stop losing /goal on daemon resume by @wenshao in https://github.com/QwenLM/qwen-code/pull/6561 * feat(core): improve subagent delegation defaults and guardrails by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7048 * feat(review): build every Step 5 auditor of a round in one call, identity first by @wenshao in https://github.com/QwenLM/qwen-code/pull/7150 * feat(channels): observe group names from inbound messages by @BenGuanRan in https://github.com/QwenLM/qwen-code/pull/7155 * fix(web-shell): recover new-session decisions wrapped in prose or fences by @wenshao in https://github.com/QwenLM/qwen-code/pull/7122 * feat(web-shell): git status chip, visual working-tree diff, and sidebar git status by @wenshao in https://github.com/QwenLM/qwen-code/pull/7054 * feat(channels): recall relevant memory per message by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7157 * ci(autofix): fan out review targets and stop route-scan starvation by @wenshao in https://github.com/QwenLM/qwen-code/pull/7127 * fix(web-shell): make approval and question overlays keyboard accessible by @wenshao in https://github.com/QwenLM/qwen-code/pull/7074 * fix(cli): restore cancelled prompt after streamed output by @barry166 in https://github.com/QwenLM/qwen-code/pull/7149 * ci(shepherd): add Fleet Shepherd — automated unblocking of the bot-PR fleet by @wenshao in https://github.com/QwenLM/qwen-code/pull/7142 * fix(channels): scope pairing and allowlist state by workspace by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7065 * refactor(core): Classify shell safety as read-only, write, or unknown by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7053 * chore(vscode-ide-companion): sync third-party notices and guard against future drift by @wenshao in https://github.com/QwenLM/qwen-code/pull/7161 * feat(cli): Add bounded daemon log rotation by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/6969 * fix(sdk-java): preserve exception cause in AcpClient init failures by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7189 * fix(web-shell): proxy /goals route in Vite dev server by @wenshao in https://github.com/QwenLM/qwen-code/pull/7187 * fix(sdk-java): correct TIMEOUT_30_MINUTES to actually be 30 minutes by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7188 * fix(cli): emit deferred stream-json startup warnings by @barry166 in https://github.com/QwenLM/qwen-code/pull/7174 * fix(review): judge a CI check by its name's latest run, not by any leftover by @wenshao in https://github.com/QwenLM/qwen-code/pull/7183 * fix(review): count Step 6's inline findings from the drafted comments, never from typed numbers by @wenshao in https://github.com/QwenLM/qwen-code/pull/7173 * feat(review): bake the round label into a findings role's identity line and key by @wenshao in https://github.com/QwenLM/qwen-code/pull/7171 * perf(channels): cache channel memory recall by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7175 * fix(web-shell): dedupe restored images and harden the sidebar shortcut handler by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7169 * feat(webshell): replay ChatRecord history in readonly WebShell by @cxruan in https://github.com/QwenLM/qwen-code/pull/6999 * perf(cli): Defer TUI runtime from ACP startup by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7182 * fix(mcp): normalize tool names for strict providers by @ran411285752 in https://github.com/QwenLM/qwen-code/pull/6976 * ci(autofix): harden the address path against stale targets and untrusted route events by @wenshao in https://github.com/QwenLM/qwen-code/pull/7163 * fix(core): apply native tool calling schema for gemma 4 by @ghisguth in https://github.com/QwenLM/qwen-code/pull/7177 * fix(cli): share one process.stdout resize listener in useTerminalSize by @mvanhorn in https://github.com/QwenLM/qwen-code/pull/7186 * feat(ui): let the user read the full plan from the exit_plan_mode confirmation by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7060 * fix: surface underlying .cause of OpenAI-compatible connection errors in debug log and API error message by @mvanhorn in https://github.com/QwenLM/qwen-code/pull/7010 * fix(web-shell): prevent toolbar label clipping by @carffuca in https://github.com/QwenLM/qwen-code/pull/7196 * fix(review): one disclosure per subject — dedupe Not-reviewed, collapse an all-built-none-launched roster by @wenshao in https://github.com/QwenLM/qwen-code/pull/7190 * fix(review): an unverified Critical must not become a public blocker — soften the Request changes it rides by @wenshao in https://github.com/QwenLM/qwen-code/pull/7191 * feat(providers): add qwen3.8-max-preview to Token Plan model list by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7199 * feat(autofix): label-driven takeover and release; fix forced-dispatch green no-op by @wenshao in https://github.com/QwenLM/qwen-code/pull/7165 * fix(autofix): stage SKILL.md beside run-agent.mjs so review-address boots (P0, regression from #7165) by @wenshao in https://github.com/QwenLM/qwen-code/pull/7225 * fix(ci): consolidate issue triage ownership by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7180 * fix(cli): allow goal controls during active loops by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7202 * test: raise timeout ceiling for I/O-bound tests flaky under CI contention by @wenshao in https://github.com/QwenLM/qwen-code/pull/7230 * fix(cli): show mode indicator alongside steering hint during streaming by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7219 * fix(scripts): allow multiple dev:daemon instances by probing Vite port by @wenshao in https://github.com/QwenLM/qwen-code/pull/7212 * feat(autofix): direct takeover of maintainer-fork PRs by @wenshao in https://github.com/QwenLM/qwen-code/pull/7213 * feat(daemon): Advertise ACP preheat readiness by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7200 * feat(autofix): surface the running model in every autofix report by @wenshao in https://github.com/QwenLM/qwen-code/pull/7226 * fix(review): say an unread chunk once, under its cause by @wenshao in https://github.com/QwenLM/qwen-code/pull/7234 * fix(autofix): a no-output crash must not advance the review watermark by @wenshao in https://github.com/QwenLM/qwen-code/pull/7229 * fix(review): prohibit isolation param in roster output and SKILL.md by @wenshao in https://github.com/QwenLM/qwen-code/pull/7235 * feat(review): retry transient API failures once; surface quota clearly by @wenshao in https://github.com/QwenLM/qwen-code/pull/7233 * feat(core): inspect persisted conversation branches by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7185 * fix(cli): align npm update checks with global registry by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7224 * feat(core): Route Plan-mode shell commands by safety by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7172 * test(autofix): exercise the SKILL stage↔resolve contract end-to-end by @wenshao in https://github.com/QwenLM/qwen-code/pull/7227 * feat(ci): auto-open a deflake fix issue for confirmed flaky tests by @wenshao in https://github.com/QwenLM/qwen-code/pull/7231 * feat(autofix): auto-manage the bot's own fork PRs without a label by @wenshao in https://github.com/QwenLM/qwen-code/pull/7243 * feat(web-shell): worktree-isolated sessions for parallel tasks by @wenshao in https://github.com/QwenLM/qwen-code/pull/7221 * feat(i18n): update catalan translaiton by @jordimas in https://github.com/QwenLM/qwen-code/pull/7253 * fix(test): deflake tool-control E2E content assertions by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7261 * fix(sdk): abort SSE request on iterator exit to release daemon subscriber by @chinesepowered in https://github.com/QwenLM/qwen-code/pull/7257 * fix(core): Enforce Plan mode entry boundary by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7248 * test(channels): add memory recall evaluation baseline by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7220 * fix(channels): exclude discrete ACP messages from replies by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7223 * fix(ci): create remote-tracking ref for fork PRs in autofix verify gate by @wenshao in https://github.com/QwenLM/qwen-code/pull/7281 * fix(sdk): clean up SSE requests on errors and dispose by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7269 * test(autofix): sync workflow assertions with split model vars by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7297 * fix(cli): repaint the TUI after OS sleep/wake or SIGCONT by @wenshao in https://github.com/QwenLM/qwen-code/pull/7265 * fix(mobile-mcp): restore bounds in UI hierarchy dumps by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/7321 * fix(test): stabilize 3 flaky tests (serve startup, usage-stats TTL, TasksStatusMessage keypress) by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7319 * docs(cli): remove stale include directories limit by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/7326 * test(feishu): enforce markdown chunk limit assertion by @ZijianZhang989 in https://github.com/QwenLM/qwen-code/pull/7324 * fix(cli): Preserve cancellation during permission prompts by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7295 * feat: add CODEOWNERS for core harness modules by @pomelo-nwu in https://github.com/QwenLM/qwen-code/pull/7304 * feat(web-shell): support custom slash command actions by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7267 * fix(core): estimate reasoning_tokens when completion_tokens_details is missing by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7239 * fix(ci): tighten API error detection to avoid false positive on review prose by @wenshao in https://github.com/QwenLM/qwen-code/pull/7328 * fix(web-shell): localize extension tag labels by @callmeYe in https://github.com/QwenLM/qwen-code/pull/7337 * fix(test): stabilize list_directory E2E — accept text output when model skips tool call by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7342 * fix(autofix): resolve owning package for nested paths; report verify-failed handoffs as not pushed by @wenshao in https://github.com/QwenLM/qwen-code/pull/7330 * fix(review): retry gh CLI on transient GitHub errors (5xx) and keyring failures by @wenshao in https://github.com/QwenLM/qwen-code/pull/7291 * feat(web-shell): add git commit history browser by @wenshao in https://github.com/QwenLM/qwen-code/pull/7204 * feat: support workspace display names by @samuelhsin in https://github.com/QwenLM/qwen-code/pull/7179 * feat(channels): add content-safe memory recall telemetry by @qqqys in https://github.com/QwenLM/qwen-code/pull/7338 * feat(daemon): restore worktree isolation on session load/resume by @wenshao in https://github.com/QwenLM/qwen-code/pull/7262 * fix(channels): deliver background agent replies by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7336 * fix(web-shell): proxy extension APIs without intercepting modules by @ytahdn in https://github.com/QwenLM/qwen-code/pull/7294 * fix(dingtalk): retry transient emotion failures by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7329 * fix(web-shell): respect voice enabled setting by @callmeYe in https://github.com/QwenLM/qwen-code/pull/7345 * feat(autofix): re-arm a stranded PR with @qwen-code /retry instead of deleting a marker by @wenshao in https://github.com/QwenLM/qwen-code/pull/7354 * fix(cli): map positional args to optional MCP prompt parameters by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7317 * fix(cli): include typed directory in /cd tab completion by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7320 * feat(serve): make ACP initialize handshake timeout configurable by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7246 * ci: move release-note classifier from per-PR workflow to release-time batch by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7339 * fix(review): make agent launches and cleanup resilient by @wenshao in https://github.com/QwenLM/qwen-code/pull/7259 * chore: add CODEOWNERS for cua-driver and mobile-mcp by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7369 * feat(autofix): pick up managed fork PRs in real time instead of waiting for the throttled schedule by @wenshao in https://github.com/QwenLM/qwen-code/pull/7350 * fix(ci): stop a slow patrol classifier from killing every flaky rerun by @wenshao in https://github.com/QwenLM/qwen-code/pull/7358 * chore: simplify CODEOWNERS to package-level rules by @pomelo-nwu in https://github.com/QwenLM/qwen-code/pull/7376 * fix(ci): tell a visuals coverage gap apart from "no visual change" by @wenshao in https://github.com/QwenLM/qwen-code/pull/7375 * feat(autofix): resolve the review threads whose findings it implemented by @wenshao in https://github.com/QwenLM/qwen-code/pull/7364 * fix: ask when Auto Mode classifier is unavailable by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/7331 * fix(autofix): retry a verification-gate crash instead of burying the agent's fix by @wenshao in https://github.com/QwenLM/qwen-code/pull/7351 * fix(core): validate goal judge terminal evidence by @qwen-code-dev-bot in https://github.com/QwenLM/qwen-code/pull/7208 * fix(core): hide interaction tools in plain headless mode by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7285 * fix(cli): show worktree branch in status line instead of workspace branch by @wenshao in https://github.com/QwenLM/qwen-code/pull/7367 * fix(acp-bridge): make detachClient idempotent via per-clientId attach-ref ledger by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7386 * feat(autofix): render the managed fleet into the scan's run summary by @wenshao in https://github.com/QwenLM/qwen-code/pull/7355 * fix(autofix): refuse a non-main takeover out loud instead of only in the job log by @wenshao in https://github.com/QwenLM/qwen-code/pull/7382 * fix(cli): update npm installs safely in background by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7322 * fix(core): support qwen3.8 side queries on DashScope by @yiliang114 in https://github.com/QwenLM/qwen-code/pull/7303 * fix(autofix): retry a model API error instead of stranding the PR by @wenshao in https://github.com/QwenLM/qwen-code/pull/7247 * fix(ci): stop /resolve reports from being guillotined mid-sentence by @wenshao in https://github.com/QwenLM/qwen-code/pull/7389 * feat(core): add fork_turns to fork subagents by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7346 * fix(web-shell): restore scheduled task reference interactions by @BZ-D in https://github.com/QwenLM/qwen-code/pull/7313 * feat(auth): add Singapore Token Plan region by @han-dreamer in https://github.com/QwenLM/qwen-code/pull/7280 * fix(core): bound web-fetch post-processing and preserve fetched content on failure by @tanzhenxin in https://github.com/QwenLM/qwen-code/pull/7305 * perf(autofix): raise fleet simultaneity from 3 to 5 by @wenshao in https://github.com/QwenLM/qwen-code/pull/7396 * perf(telemetry): lazy-load the SDK and split OTLP exporter chains by protocol by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7276 * fix(core): Fence concurrent ACP session writers by @doudouOUC in https://github.com/QwenLM/qwen-code/pull/7237 * feat(web-shell): surface worktree isolation in the new-session empty state by @wenshao in https://github.com/QwenLM/qwen-code/pull/7365 * fix(ci): serialise the two workflows that push to a PR head branch by @wenshao in https://github.com/QwenLM/qwen-code/pull/7392 * fix(web-shell): persist the daemon bearer token per-tab so it survives refresh by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7374 * fix(core,cli): drain background notifications outside the subagent's ALS frame by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7194 * fix(acp-bridge): map Windows-shaped workspace paths to their sandbox mount by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7228 * fix(core): salvage session usage into the history before deleting transcripts by @zjunothing in https://github.com/QwenLM/qwen-code/pull/7391 * fix(core): clarify background agent continuation by @DragonnZhang in https://github.com/QwenLM/qwen-code/pull/7300 * fix(transcript): mark dangling tool history incomplete by @cxruan in https://github.com/QwenLM/qwen-code/pull/7340 * fix(mcp): add opt-in model payload filtering by @LaZzyMan in https://github.com/QwenLM/qwen-code/pull/7413 ## New Contributors * @xurik made their first contribution in https://github.com/QwenLM/qwen-code/pull/6991 * @azurecgx made their first contribution in https://github.com/QwenLM/qwen-code/pull/6957 * @makwingchi made their first contribution in https://github.com/QwenLM/qwen-code/pull/7082 * @ARE404 made their first contribution in https://github.com/QwenLM/qwen-code/pull/7104 * @ghisguth made their first contribution in https://github.com/QwenLM/qwen-code/pull/7177 **Full Changelog**: https://github.com/QwenLM/qwen-code/compare/cua-driver-rs-v0.7.2...cua-driver-rs-v0.7.3