mirror of
https://github.com/QwenLM/qwen-code.git
synced 2026-07-09 17:19:02 +00:00
docs(changelog): sync for v0.19.7
This commit is contained in:
github-actions[bot]
parent
47978cd9e4
commit
dbb12f32d0
1 changed files with 158 additions and 0 deletions
158
CHANGELOG.md
158
CHANGELOG.md
|
|
@ -12,6 +12,164 @@ are listed; nightly and preview pre-releases are intentionally omitted.
|
|||
> [GitHub Releases](https://github.com/QwenLM/qwen-code/releases). Do not edit it
|
||||
> by hand — run `npm run changelog` to regenerate.
|
||||
|
||||
## [0.19.7](https://github.com/QwenLM/qwen-code/releases/tag/v0.19.7) - 2026-07-07
|
||||
|
||||
### Added
|
||||
|
||||
- review: route suggestion-level findings to an updatable PR comment ([#5786](https://github.com/QwenLM/qwen-code/pull/5786))
|
||||
- serve: Add runtime.activity fields to daemon status API ([#6270](https://github.com/QwenLM/qwen-code/pull/6270))
|
||||
- web-shell: add a daemon status page backed by GET /daemon/status ([#6272](https://github.com/QwenLM/qwen-code/pull/6272))
|
||||
- web-shell: add MCP mentions and iconized @ references ([#6279](https://github.com/QwenLM/qwen-code/pull/6279))
|
||||
- web-shell: manage sessions from the sidebar (archive, unarchive, delete) ([#6293](https://github.com/QwenLM/qwen-code/pull/6293))
|
||||
- daemon: Add session export endpoint ([#6297](https://github.com/QwenLM/qwen-code/pull/6297))
|
||||
- acp: advertise vision-bridge image capability in initialize response ([#6269](https://github.com/QwenLM/qwen-code/pull/6269))
|
||||
- web-shell: support compact echarts full data blocks ([#6232](https://github.com/QwenLM/qwen-code/pull/6232))
|
||||
- acp: Batch session load replay ([#6309](https://github.com/QwenLM/qwen-code/pull/6309))
|
||||
- web-shell: add custom at mention panel ([#6242](https://github.com/QwenLM/qwen-code/pull/6242))
|
||||
- acp-bridge: Add EventBus subscriber byte cap ([#6314](https://github.com/QwenLM/qwen-code/pull/6314))
|
||||
- web-shell: time-series metrics charts on Daemon Status ([#6307](https://github.com/QwenLM/qwen-code/pull/6307))
|
||||
- daemon: Add session organization ([#6305](https://github.com/QwenLM/qwen-code/pull/6305))
|
||||
- cli: Surface daemon prompt queue status ([#6325](https://github.com/QwenLM/qwen-code/pull/6325))
|
||||
- scheduler: opt-in per-tool-call execution timeout ([#6124](https://github.com/QwenLM/qwen-code/pull/6124))
|
||||
- web-shell: add onSessionChange and onSubmitBefore callbacks ([#6333](https://github.com/QwenLM/qwen-code/pull/6333))
|
||||
- core: stabilize tool schema declaration order ([#6339](https://github.com/QwenLM/qwen-code/pull/6339))
|
||||
- core: model fallback chain — auto-switch to backup models on overload ([#6273](https://github.com/QwenLM/qwen-code/pull/6273))
|
||||
- cli: support multi-folder workspaces in file system boundary checks ([#6278](https://github.com/QwenLM/qwen-code/pull/6278))
|
||||
- web-shell: support icon chips for mention tags ([#6337](https://github.com/QwenLM/qwen-code/pull/6337))
|
||||
- LSP Server support hot reload ([#5953](https://github.com/QwenLM/qwen-code/pull/5953))
|
||||
- cli: Add large pipe frame measurement ([#6335](https://github.com/QwenLM/qwen-code/pull/6335))
|
||||
- web-shell: named session groups and color tags in the sidebar ([#6350](https://github.com/QwenLM/qwen-code/pull/6350))
|
||||
- web-shell: add a Scheduled Tasks management page ([#6348](https://github.com/QwenLM/qwen-code/pull/6348))
|
||||
- web-shell: show Settings and Daemon Status as an in-place panel ([#6341](https://github.com/QwenLM/qwen-code/pull/6341))
|
||||
- web-shell: add token-usage analytics dashboard to Daemon Status ([#6388](https://github.com/QwenLM/qwen-code/pull/6388))
|
||||
- cli: Add Phase 1 workspace runtime registry ([#6394](https://github.com/QwenLM/qwen-code/pull/6394))
|
||||
- core: surface PreToolUse hook 'ask' as a TUI confirmation ([#5629](https://github.com/QwenLM/qwen-code/pull/5629))
|
||||
- review: add issue-fidelity and root-cause ownership gate to /review ([#6395](https://github.com/QwenLM/qwen-code/pull/6395))
|
||||
- cli: Add Phase 2a workspace foundation ([#6410](https://github.com/QwenLM/qwen-code/pull/6410))
|
||||
- web-shell: add Session Overview panel and in-window split view ([#6400](https://github.com/QwenLM/qwen-code/pull/6400))
|
||||
- cli: add --project and --global flags to /model for per-project model persistence ([#6060](https://github.com/QwenLM/qwen-code/pull/6060))
|
||||
- core: add maxSubAgents setting to limit parallel sub-agent count ([#6354](https://github.com/QwenLM/qwen-code/pull/6354))
|
||||
- scheduled-tasks: run each task in its own dedicated, named session ([#6389](https://github.com/QwenLM/qwen-code/pull/6389))
|
||||
- cli: support stacked slash-skill invocations ([#6361](https://github.com/QwenLM/qwen-code/pull/6361))
|
||||
- core: add Tool(param:value) permission syntax for parameter-level access control ([#6106](https://github.com/QwenLM/qwen-code/pull/6106))
|
||||
- core: add tools.visible config for selective deferred-tool visibility at startup ([#6372](https://github.com/QwenLM/qwen-code/pull/6372))
|
||||
- web-shell: add Qwen logo beside the sidebar new-chat button ([#6437](https://github.com/QwenLM/qwen-code/pull/6437))
|
||||
- web-shell: add column reorder, resize, and freeze controls to markdown table ([#6444](https://github.com/QwenLM/qwen-code/pull/6444))
|
||||
- channels: add WeCom intelligent robot channel ([#6436](https://github.com/QwenLM/qwen-code/pull/6436))
|
||||
- web-shell: unify scheduled task sessions — bind chat-created tasks + clock icon ([#6453](https://github.com/QwenLM/qwen-code/pull/6453))
|
||||
|
||||
### Changed
|
||||
|
||||
- core: centralize extension runtime refresh ([#6152](https://github.com/QwenLM/qwen-code/pull/6152))
|
||||
|
||||
### Fixed
|
||||
|
||||
- triage: strengthen PR gate with batch detection, problem existence check, and red flag patterns ([#5723](https://github.com/QwenLM/qwen-code/pull/5723))
|
||||
- autofix: unconditionally restore tracked files before branch checkout (#6281) ([#6286](https://github.com/QwenLM/qwen-code/pull/6286))
|
||||
- vscode: keep auth quick inputs open on focus loss ([#6274](https://github.com/QwenLM/qwen-code/pull/6274))
|
||||
- cache: preserve tools prefix in side-query for Anthropic prompt-cache hits ([#6225](https://github.com/QwenLM/qwen-code/pull/6225))
|
||||
- core: give Stop-hook continuations a fresh per-turn tool-call budget; make the cap configurable ([#6238](https://github.com/QwenLM/qwen-code/pull/6238))
|
||||
- web-shell: use theme color for @ group titles ([#6294](https://github.com/QwenLM/qwen-code/pull/6294))
|
||||
- acp: pass per-session settings explicitly instead of racing on this.settings ([#6292](https://github.com/QwenLM/qwen-code/pull/6292))
|
||||
- core: improve debug txt diagnostics ([#6277](https://github.com/QwenLM/qwen-code/pull/6277))
|
||||
- ci: require maintainer-applied `autofix/approved` label for tier-1 fast-path ([#6276](https://github.com/QwenLM/qwen-code/pull/6276))
|
||||
- auth: prevent persistent 401 after API key change ([#6284](https://github.com/QwenLM/qwen-code/pull/6284))
|
||||
- cli: stream long responses into scrollback to stop scroll-to-top lock ([#6170](https://github.com/QwenLM/qwen-code/pull/6170))
|
||||
- qqbot: streaming idle-flush with tool-call and stale-callback protection ([#6204](https://github.com/QwenLM/qwen-code/pull/6204))
|
||||
- openai: preserve descriptionless tools ([#6243](https://github.com/QwenLM/qwen-code/pull/6243))
|
||||
- core: enforce agent concurrency cap on foreground sub-agents ([#6300](https://github.com/QwenLM/qwen-code/pull/6300))
|
||||
- ci: Stop review bots for closed PRs ([#6304](https://github.com/QwenLM/qwen-code/pull/6304))
|
||||
- core: skip abbreviations in multiple_sentences filter (#6077) ([#6193](https://github.com/QwenLM/qwen-code/pull/6193))
|
||||
- ci: skip stale PR review runs ([#6313](https://github.com/QwenLM/qwen-code/pull/6313))
|
||||
- core: treat request timeout of 0 as disabled instead of aborting immediately ([#6288](https://github.com/QwenLM/qwen-code/pull/6288))
|
||||
- serve: resolve false auth warning in preflight when API key is set via settings ([#6296](https://github.com/QwenLM/qwen-code/pull/6296))
|
||||
- core: treat @-attached files as read for prior-read enforcement ([#6295](https://github.com/QwenLM/qwen-code/pull/6295))
|
||||
- cli: preserve partial remote input JSONL records ([#6317](https://github.com/QwenLM/qwen-code/pull/6317))
|
||||
- core: disable qwen thinking via chat_template_kwargs on non-DashScope servers ([#6271](https://github.com/QwenLM/qwen-code/pull/6271))
|
||||
- web-shell: keep skill slash commands after starting a new session ([#6319](https://github.com/QwenLM/qwen-code/pull/6319))
|
||||
- web-shell: localize built-in command and skill descriptions in the slash menu ([#6326](https://github.com/QwenLM/qwen-code/pull/6326))
|
||||
- desktop: enforce transform_data isolation ([#6285](https://github.com/QwenLM/qwen-code/pull/6285))
|
||||
- core: skip no-op max_tokens escalation ([#6234](https://github.com/QwenLM/qwen-code/pull/6234))
|
||||
- core: avoid null OpenAPI schema types ([#6323](https://github.com/QwenLM/qwen-code/pull/6323))
|
||||
- core: preserve OpenAI reasoning as raw thoughts ([#6192](https://github.com/QwenLM/qwen-code/pull/6192))
|
||||
- core: add UTF-8 prefix for cmd.exe on Windows ([#6216](https://github.com/QwenLM/qwen-code/pull/6216))
|
||||
- cli: allow queued input during compression ([#6336](https://github.com/QwenLM/qwen-code/pull/6336))
|
||||
- web-shell: finalize deferred gated submissions ([#6342](https://github.com/QwenLM/qwen-code/pull/6342))
|
||||
- cli: smoother live streaming preview — drop "generating more" cue, hold back partial table rows ([#6340](https://github.com/QwenLM/qwen-code/pull/6340))
|
||||
- desktop: preserve glued automation history records ([#6344](https://github.com/QwenLM/qwen-code/pull/6344))
|
||||
- web-shell: suppress stale pending prompt refresh errors ([#6352](https://github.com/QwenLM/qwen-code/pull/6352))
|
||||
- web-shell: constrain virtual scroll rows ([#6362](https://github.com/QwenLM/qwen-code/pull/6362))
|
||||
- cli: Allow ACP local fallback reads from /tmp ([#6370](https://github.com/QwenLM/qwen-code/pull/6370))
|
||||
- core: default context windows to 200k ([#6387](https://github.com/QwenLM/qwen-code/pull/6387))
|
||||
- cli: Keep model picker entries contiguous in short terminals ([#6359](https://github.com/QwenLM/qwen-code/pull/6359))
|
||||
- core: Include request IDs in OpenAI error logs ([#6379](https://github.com/QwenLM/qwen-code/pull/6379))
|
||||
- cli: ignore current review run in presubmit CI ([#6397](https://github.com/QwenLM/qwen-code/pull/6397))
|
||||
- autofix: improve review addressing and verification ([#6382](https://github.com/QwenLM/qwen-code/pull/6382))
|
||||
- core: require integer ReadFile pagination params ([#6381](https://github.com/QwenLM/qwen-code/pull/6381))
|
||||
- core: resolve symlinks when matching conditional rules and skills ([#6371](https://github.com/QwenLM/qwen-code/pull/6371))
|
||||
- web-shell: refine tool detail presentation ([#6399](https://github.com/QwenLM/qwen-code/pull/6399))
|
||||
- core: preserve no-argument tool calls that stream an empty arguments string ([#6250](https://github.com/QwenLM/qwen-code/pull/6250))
|
||||
- cli: use EnvHttpProxyAgent in channel proxy to respect NO_PROXY (#6401) ([#6405](https://github.com/QwenLM/qwen-code/pull/6405))
|
||||
- daemon: Handle settings reload events outside transcript ([#6407](https://github.com/QwenLM/qwen-code/pull/6407))
|
||||
- memory: don't advance AutoMemory extract cursor when the agent makes zero tool calls ([#6398](https://github.com/QwenLM/qwen-code/pull/6398))
|
||||
- core: gate image payload replacement behind threshold ([#6380](https://github.com/QwenLM/qwen-code/pull/6380))
|
||||
- review: remove qwen-code-specific core-infra gate from bundled /review ([#6412](https://github.com/QwenLM/qwen-code/pull/6412))
|
||||
- web-shell: polish scheduled task timeline UI ([#6386](https://github.com/QwenLM/qwen-code/pull/6386))
|
||||
- cli: smoother streaming table rendering ([#6345](https://github.com/QwenLM/qwen-code/pull/6345))
|
||||
- core: Gate large PDF text extraction ([#6409](https://github.com/QwenLM/qwen-code/pull/6409))
|
||||
- core: allow rewind after compressed history ([#6358](https://github.com/QwenLM/qwen-code/pull/6358))
|
||||
- core: align monitor limit parameter schemas ([#6413](https://github.com/QwenLM/qwen-code/pull/6413))
|
||||
- core: prevent KV-cache invalidation on tool_search by reordering reminderParts ([#6420](https://github.com/QwenLM/qwen-code/pull/6420))
|
||||
- shell: avoid Unix pager default on Windows ([#6390](https://github.com/QwenLM/qwen-code/pull/6390))
|
||||
- daemon: preserve user message source metadata ([#6385](https://github.com/QwenLM/qwen-code/pull/6385))
|
||||
- core: prevent re-invoking loaded skill from appending duplicate content ([#6430](https://github.com/QwenLM/qwen-code/pull/6430))
|
||||
- web-shell: keep split-view session list fresh and preserve panes across view switches ([#6418](https://github.com/QwenLM/qwen-code/pull/6418))
|
||||
- web-shell: Improve user tags and mobile menu layout ([#6441](https://github.com/QwenLM/qwen-code/pull/6441))
|
||||
- web-shell: keep errored turns expanded ([#6424](https://github.com/QwenLM/qwen-code/pull/6424))
|
||||
- web-shell: clear stale floating todos ([#6425](https://github.com/QwenLM/qwen-code/pull/6425))
|
||||
- web-shell: hide rotating loading phrase in split-view pane status ([#6447](https://github.com/QwenLM/qwen-code/pull/6447))
|
||||
- core: Support large text range reads ([#6404](https://github.com/QwenLM/qwen-code/pull/6404))
|
||||
- core: strip system-reminder blocks from session title and recap side-query prompts ([#6435](https://github.com/QwenLM/qwen-code/pull/6435))
|
||||
- autofix: report review handoff failures ([#6415](https://github.com/QwenLM/qwen-code/pull/6415))
|
||||
- monitor: preserve inherited git pager ([#6429](https://github.com/QwenLM/qwen-code/pull/6429))
|
||||
- serve: classify interrupted model stream errors ([#6422](https://github.com/QwenLM/qwen-code/pull/6422))
|
||||
- web-shell: refine tool call summaries ([#6450](https://github.com/QwenLM/qwen-code/pull/6450))
|
||||
- web-shell: split-view pane fixes (remove "current" badge, clear composer on send) ([#6454](https://github.com/QwenLM/qwen-code/pull/6454))
|
||||
|
||||
### Performance
|
||||
|
||||
- glob: prune ignored directories during traversal, not just post-filter ([#6123](https://github.com/QwenLM/qwen-code/pull/6123))
|
||||
- cli: cache LoadedSettings per workspace with stat-based invalidation ([#6310](https://github.com/QwenLM/qwen-code/pull/6310))
|
||||
- ci: optimize autofix pipeline — fast-track, skip duplicate build, scoped tests ([#6315](https://github.com/QwenLM/qwen-code/pull/6315))
|
||||
- memoize skill scans, debounce sleep-inhibitor log, guard IDE readdir ([#6155](https://github.com/QwenLM/qwen-code/pull/6155))
|
||||
- core: Add session start profiler ([#6349](https://github.com/QwenLM/qwen-code/pull/6349))
|
||||
- cli: defer startup prefetch tasks ([#6303](https://github.com/QwenLM/qwen-code/pull/6303))
|
||||
|
||||
### Documentation
|
||||
|
||||
- design: daemon side-channel coordination (A1/A2/A4/A5) ([#4511](https://github.com/QwenLM/qwen-code/pull/4511))
|
||||
- fix skill invocation syntax and include Feishu in channel lists ([#6320](https://github.com/QwenLM/qwen-code/pull/6320))
|
||||
- fix settings.json reference drift against schema ([#6351](https://github.com/QwenLM/qwen-code/pull/6351))
|
||||
- web-shell: document chart renderer integration ([#6353](https://github.com/QwenLM/qwen-code/pull/6353))
|
||||
- document PreToolUse hook permissionDecision "ask" behavior ([#6411](https://github.com/QwenLM/qwen-code/pull/6411))
|
||||
- consolidate design docs and plans under docs/ ([#6417](https://github.com/QwenLM/qwen-code/pull/6417))
|
||||
|
||||
### Other
|
||||
|
||||
- ci(audio): clarify macOS prebuild artifact suffix ([#6275](https://github.com/QwenLM/qwen-code/pull/6275))
|
||||
- test(e2e): make fake OpenAI reachable from Docker sandbox ([#6302](https://github.com/QwenLM/qwen-code/pull/6302))
|
||||
- test(core): cover full:false branch of recordAttachedFileRead for truncated @-attachments ([#6324](https://github.com/QwenLM/qwen-code/pull/6324))
|
||||
- Notify model when extension capabilities change ([#6245](https://github.com/QwenLM/qwen-code/pull/6245))
|
||||
- Restart stalled ACP bridge for channels ([#6330](https://github.com/QwenLM/qwen-code/pull/6330))
|
||||
- Fix incorrect context window calculation for custom models ([#6266](https://github.com/QwenLM/qwen-code/pull/6266))
|
||||
- [codex] add proactive channel loop tools ([#6287](https://github.com/QwenLM/qwen-code/pull/6287))
|
||||
- ci(autofix): move agent prompts into a project skill ([#6306](https://github.com/QwenLM/qwen-code/pull/6306))
|
||||
- test(core): keep context warning test aligned with default token limit ([#6391](https://github.com/QwenLM/qwen-code/pull/6391))
|
||||
- Handle missing web-shell sessions without redirecting ([#6357](https://github.com/QwenLM/qwen-code/pull/6357))
|
||||
- Avoid refreshing session activity on load ([#6439](https://github.com/QwenLM/qwen-code/pull/6439))
|
||||
- Upgrade GitHub Actions for Node 24 compatibility ([#5157](https://github.com/QwenLM/qwen-code/pull/5157))
|
||||
- [codex] add natural channel memory intents ([#6376](https://github.com/QwenLM/qwen-code/pull/6376))
|
||||
|
||||
## [0.19.6](https://github.com/QwenLM/qwen-code/releases/tag/v0.19.6) - 2026-07-03
|
||||
|
||||
### Added
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue