@qwen-code/qwen-code 0.19.3 → 0.19.4-nightly.20260702.46814e4f1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (308) hide show
  1. package/bundled/loop/SKILL.md +22 -3
  2. package/bundled/loop/SKILL.test.ts +20 -0
  3. package/bundled/loop/autonomous-loop.test.ts +59 -0
  4. package/bundled/loop/autonomous-loop.ts +102 -0
  5. package/bundled/loop/loop-task-file.test.ts +1064 -0
  6. package/bundled/loop/loop-task-file.ts +404 -0
  7. package/bundled/loop/loop-tick-resolver.test.ts +906 -0
  8. package/bundled/loop/loop-tick-resolver.ts +416 -0
  9. package/bundled/qc-helper/SKILL.md +12 -5
  10. package/bundled/qc-helper/docs/configuration/auth.md +1 -1
  11. package/bundled/qc-helper/docs/configuration/settings.md +62 -32
  12. package/bundled/qc-helper/docs/features/approval-mode.md +3 -0
  13. package/bundled/qc-helper/docs/features/auto-mode.md +40 -0
  14. package/bundled/qc-helper/docs/features/channels/overview.md +81 -18
  15. package/bundled/qc-helper/docs/features/channels/plugins.md +1 -1
  16. package/bundled/qc-helper/docs/features/code-review.md +17 -61
  17. package/bundled/qc-helper/docs/features/commands.md +42 -40
  18. package/bundled/qc-helper/docs/features/headless.md +19 -18
  19. package/bundled/qc-helper/docs/features/hooks.md +6 -1
  20. package/bundled/qc-helper/docs/features/mcp.md +3 -1
  21. package/bundled/qc-helper/docs/features/memory.md +1 -1
  22. package/bundled/qc-helper/docs/features/scheduled-tasks.md +10 -0
  23. package/bundled/qc-helper/docs/features/status-line.md +3 -3
  24. package/bundled/qc-helper/docs/ide-integration/ide-integration.md +4 -4
  25. package/bundled/qc-helper/docs/qwen-serve.md +53 -0
  26. package/bundled/qc-helper/docs/support/troubleshooting.md +11 -0
  27. package/bundled/review/DESIGN.md +21 -26
  28. package/bundled/review/SKILL.md +45 -129
  29. package/chunks/MaxSizedBox-GHGRQ2OS.js +69 -0
  30. package/chunks/{StandaloneSessionPicker-Z52JTYST.js → StandaloneSessionPicker-G7YTTM27.js} +49 -42
  31. package/chunks/{acpAgent-7RZUOWOM.js → acpAgent-SDOILBSU.js} +1325 -365
  32. package/chunks/{agent-4FG7WTHJ.js → agent-NTSXGV4L.js} +31 -25
  33. package/chunks/agent-headless-WYURMSAB.js +62 -0
  34. package/chunks/{anthropicContentGenerator-PFITVL7Y.js → anthropicContentGenerator-WL3LJOUB.js} +110 -31
  35. package/chunks/{artifact-tool-D2X7G3C3.js → artifact-tool-W76WC2JX.js} +3 -3
  36. package/chunks/{askUserQuestion-UGETICJQ.js → askUserQuestion-J3P6TH3V.js} +2 -2
  37. package/chunks/bridge-AMOL4W4P.js +73 -0
  38. package/chunks/{ca-IUFWESZL.js → ca-HCMB2I54.js} +10 -9
  39. package/chunks/channel-worker-supervisor-C3IB63V7.js +659 -0
  40. package/chunks/{chunk-STUWFKGD.js → chunk-26JIREA5.js} +466 -543
  41. package/chunks/{chunk-SLNKLLEO.js → chunk-2MRV62AU.js} +37 -8
  42. package/chunks/{chunk-ODLIHNOY.js → chunk-2U4HUMOL.js} +1 -1
  43. package/chunks/{chunk-CKQA6AWF.js → chunk-4D73HUOP.js} +1 -1
  44. package/chunks/{chunk-M2U5NM24.js → chunk-54YZOP7P.js} +309 -72
  45. package/chunks/{chunk-J7FA27RO.js → chunk-5LKL3PAU.js} +3 -3
  46. package/chunks/chunk-5P5XGNYH.js +93 -0
  47. package/chunks/{chunk-UFLGERMB.js → chunk-6JE32BRB.js} +178 -124
  48. package/chunks/{chunk-FKAVULLM.js → chunk-75AVTQJ2.js} +74 -10
  49. package/chunks/{chunk-I2LYBGNI.js → chunk-7IOMIYOE.js} +1 -1
  50. package/chunks/{chunk-4LUZAKMK.js → chunk-7KXAGQJ2.js} +1 -1
  51. package/chunks/{chunk-BXSHL2SF.js → chunk-7LUOQJ3D.js} +1 -1
  52. package/chunks/{chunk-YAINVQ3Q.js → chunk-A5OPLHFM.js} +18693 -17055
  53. package/chunks/chunk-AHNXYU4O.js +24 -0
  54. package/chunks/{chunk-UX4SGCGN.js → chunk-AVQDCKNF.js} +96 -22
  55. package/chunks/{chunk-HYCC3XM3.js → chunk-B236XPGF.js} +1 -1
  56. package/chunks/chunk-B6SZLQAZ.js +3523 -0
  57. package/chunks/{chunk-5DWXLGOM.js → chunk-BJBWRCSK.js} +1 -1
  58. package/chunks/chunk-BPFJQGY7.js +193 -0
  59. package/chunks/{chunk-SKQZZVAV.js → chunk-BZHLWC6T.js} +2 -2
  60. package/chunks/{chunk-U3KSMM7G.js → chunk-D5KISTLW.js} +339 -383
  61. package/chunks/{chunk-GT2FGICD.js → chunk-DE7YXUR7.js} +1 -1
  62. package/chunks/{chunk-BUUQ5HPX.js → chunk-DG573EOO.js} +1 -1
  63. package/chunks/{chunk-VVLOGDB6.js → chunk-DX3TKXNT.js} +2 -2
  64. package/chunks/chunk-E5A7LHNN.js +51 -0
  65. package/chunks/{chunk-3PJXIDKI.js → chunk-E6US47KI.js} +3 -27
  66. package/chunks/{chunk-QBZ22PQX.js → chunk-EWALXUAD.js} +3 -1
  67. package/chunks/{chunk-PDE7CSDS.js → chunk-FXKI7OW6.js} +2 -2
  68. package/chunks/{chunk-F3ZYSEY6.js → chunk-FY4TT4L4.js} +3 -3
  69. package/chunks/{chunk-QLDPRILK.js → chunk-G7OQ3RFJ.js} +3 -3
  70. package/chunks/{chunk-R2BXK7SX.js → chunk-GHOTR7HL.js} +6 -1
  71. package/chunks/{chunk-KMGMKW5Q.js → chunk-GLPV75ES.js} +9 -3
  72. package/chunks/{chunk-LD2XBG6Z.js → chunk-GO6LNQXT.js} +1 -39
  73. package/chunks/{chunk-CNLEDYE3.js → chunk-H6TNLOLM.js} +5 -5
  74. package/chunks/{chunk-JVHEVCAU.js → chunk-HLW4EGKR.js} +31 -6
  75. package/chunks/{chunk-I4H7PRWQ.js → chunk-HTWHAN44.js} +1 -1
  76. package/chunks/{chunk-QE6KJYLT.js → chunk-IPYE5BIS.js} +123 -25
  77. package/chunks/{chunk-EJYJYJZ2.js → chunk-IRJWHZWU.js} +3 -70
  78. package/chunks/{chunk-EDMZZFDU.js → chunk-J3YPTTNL.js} +930 -798
  79. package/chunks/{chunk-GGBIN5K7.js → chunk-JGAJMIJZ.js} +1 -1
  80. package/chunks/{chunk-N4EJFBN5.js → chunk-JVLP5US5.js} +427 -263
  81. package/chunks/{chunk-R5OP2ATH.js → chunk-JZG7VB6C.js} +11 -11
  82. package/chunks/{chunk-QI7GOGJR.js → chunk-KKQWC6WS.js} +1 -1
  83. package/chunks/{chunk-ADCIHOQZ.js → chunk-KX5QZRW2.js} +142 -30
  84. package/chunks/{chunk-SBBN4BWR.js → chunk-L7OJ3ZO3.js} +1 -1
  85. package/chunks/{chunk-NDOC6M5S.js → chunk-L7QPM2YD.js} +2 -2
  86. package/chunks/{chunk-BVSVIN45.js → chunk-LKWAVAB5.js} +3 -3
  87. package/chunks/{chunk-T4T2FNV4.js → chunk-LOW52WYZ.js} +2 -2
  88. package/chunks/{chunk-4QK3O43S.js → chunk-LU7BIVWC.js} +4 -4
  89. package/chunks/{chunk-ZP6UKOPB.js → chunk-LWGLBVWA.js} +1 -1
  90. package/chunks/{chunk-L7OKCI7T.js → chunk-M7BLWURA.js} +15 -1
  91. package/chunks/{chunk-PR32FRT6.js → chunk-MIM62IXS.js} +3 -3
  92. package/chunks/{chunk-RD6YJJJ6.js → chunk-MOUXUCXQ.js} +1 -1
  93. package/chunks/{chunk-VBZ452JT.js → chunk-MSNGFH6L.js} +494 -114
  94. package/chunks/{chunk-BX5YO7UW.js → chunk-MTNTVOJR.js} +542 -370
  95. package/chunks/chunk-MX2YXRER.js +44 -0
  96. package/chunks/chunk-NIOPRZNN.js +27 -0
  97. package/chunks/{chunk-SMYQZVZ3.js → chunk-NOGSPTLJ.js} +1 -1
  98. package/chunks/{chunk-DDMOU62Q.js → chunk-NPQWQYU6.js} +1 -1
  99. package/chunks/{chunk-MNF6S5PR.js → chunk-P5THQV6A.js} +4 -4
  100. package/chunks/{chunk-6V4NZFUS.js → chunk-PCYOT4PK.js} +5 -5
  101. package/chunks/{chunk-CF6BB3LT.js → chunk-PPFWEV7G.js} +10 -1
  102. package/chunks/{chunk-FXKVLFFY.js → chunk-PSFDB5FZ.js} +3 -3
  103. package/chunks/{chunk-XJYUPHRC.js → chunk-QOMAHFYI.js} +3 -3
  104. package/chunks/{chunk-J37FGIOA.js → chunk-QXPI7FEC.js} +38 -4
  105. package/chunks/{chunk-RRFZXMYP.js → chunk-QZDBHNMV.js} +691 -164
  106. package/chunks/{chunk-S7AKE7HL.js → chunk-R6RI2HIC.js} +63 -9
  107. package/chunks/{chunk-HZT4UCFB.js → chunk-RB7S52NJ.js} +4 -4
  108. package/chunks/{chunk-J33ZR6OJ.js → chunk-RK22ACPP.js} +16 -16
  109. package/chunks/chunk-RM3PGSGL.js +66 -0
  110. package/chunks/chunk-RNBYOUGV.js +42 -0
  111. package/chunks/chunk-S6RLAIUR.js +23 -0
  112. package/chunks/{chunk-3TV3HUWZ.js → chunk-SCYK2YLI.js} +81 -30
  113. package/chunks/{chunk-HPZTOP4I.js → chunk-SEKRT4SD.js} +133 -23
  114. package/chunks/chunk-SEZC2725.js +36 -0
  115. package/chunks/chunk-SIUQ3YYX.js +78 -0
  116. package/chunks/{chunk-O2VBQ3Y2.js → chunk-SJBIFVD2.js} +2 -2
  117. package/chunks/chunk-SYCJMSIJ.js +82 -0
  118. package/chunks/{chunk-JE5FVRPC.js → chunk-T22H4V4R.js} +48 -23
  119. package/chunks/{chunk-3227DZO4.js → chunk-T42ACHFI.js} +1 -1
  120. package/chunks/{chunk-UO5GVXQE.js → chunk-TENOUWY2.js} +3 -3
  121. package/chunks/{chunk-JOLIREYR.js → chunk-TOAUDIS7.js} +1 -1
  122. package/chunks/{chunk-QNKBEJO2.js → chunk-UA6XBZT7.js} +2776 -1289
  123. package/chunks/{chunk-O75BHFQW.js → chunk-UQTYK5SS.js} +2 -2
  124. package/chunks/{chunk-Y6YMRS3U.js → chunk-UY2TCIZS.js} +17 -3
  125. package/chunks/{chunk-FNJ7BDPH.js → chunk-UY7ZAYTY.js} +10 -7
  126. package/chunks/{chunk-UYT3MWEB.js → chunk-UYHWM24C.js} +30 -2
  127. package/chunks/{chunk-SMXULRAN.js → chunk-VXGPYF7U.js} +1 -1
  128. package/chunks/{chunk-BTLJ2MF5.js → chunk-W7CG5HPC.js} +3 -3
  129. package/chunks/{chunk-JJXWTVSK.js → chunk-XHZKDN7M.js} +34 -1
  130. package/chunks/{chunk-K5XU6E4T.js → chunk-XKSCTXPK.js} +2 -2
  131. package/chunks/chunk-Y4LITCRB.js +367 -0
  132. package/chunks/chunk-YHTNWGRR.js +239 -0
  133. package/chunks/{chunk-BKXBARJZ.js → chunk-YPCPZSZB.js} +1 -1
  134. package/chunks/{chunk-QWFNTRH4.js → chunk-YSM24SCE.js} +1 -1
  135. package/chunks/chunk-YTFBURQD.js +313 -0
  136. package/chunks/{chunk-NL4C3H6C.js → chunk-YXK4GPYO.js} +1 -1
  137. package/chunks/{chunk-RYLJU2DY.js → chunk-ZN5T4BHI.js} +1 -1
  138. package/chunks/{computer-use-JGB36FDW.js → computer-use-MTSVPLPY.js} +34 -28
  139. package/chunks/contextCommand-VGJ3BPTK.js +67 -0
  140. package/chunks/{cron-create-SKEY55RG.js → cron-create-6YTCOX6P.js} +4 -4
  141. package/chunks/{cron-delete-I3NAT32Z.js → cron-delete-E3NIFSUK.js} +4 -4
  142. package/chunks/{cron-list-6XRRL6TH.js → cron-list-Q64MNQNE.js} +4 -4
  143. package/chunks/daemon-EEB2RYOL.js +8453 -0
  144. package/chunks/daemon-status-provider-H5SNMK42.js +71 -0
  145. package/chunks/{de-EOW5PVHF.js → de-PLLFDERI.js} +10 -9
  146. package/chunks/{dist-NHEYYB2B.js → dist-ADMIQGHI.js} +2 -2
  147. package/chunks/{dist-J37VXPLD.js → dist-BCQ7INM7.js} +40 -15
  148. package/chunks/{dist-663NZ6MI.js → dist-KNZFCY5F.js} +116 -11
  149. package/chunks/{dist-LB6OGPVM.js → dist-LHUWC3K6.js} +26 -5
  150. package/chunks/{dist-PQPZ5Q4N.js → dist-WBBZ76JZ.js} +20 -9
  151. package/chunks/{earlyInputCapture-44RISVVJ.js → earlyInputCapture-U4ZIQSQU.js} +32 -26
  152. package/chunks/{edit-NN6GM5DT.js → edit-SYIQ2KBI.js} +41 -38
  153. package/chunks/{en-HPWEJBTR.js → en-M4EBUBNU.js} +32 -10
  154. package/chunks/{enter-worktree-PB3O2C7D.js → enter-worktree-HPPOO244.js} +31 -25
  155. package/chunks/{enterPlanMode-YQFNNSUA.js → enterPlanMode-56YYO52U.js} +41 -25
  156. package/chunks/{errors-LS7XU5DZ.js → errors-EO4SXCFP.js} +33 -27
  157. package/chunks/{exit-worktree-E2FTNSBK.js → exit-worktree-UDM7E6XH.js} +31 -25
  158. package/chunks/{exitPlanMode-GLYK2BUB.js → exitPlanMode-M6P3AWLJ.js} +149 -33
  159. package/chunks/{fast-path-WQ6M7LI7.js → fast-path-42P4QYWI.js} +33 -14
  160. package/chunks/{fast-path-settings-OAGU3X5H.js → fast-path-settings-UQLRFDQP.js} +2 -2
  161. package/chunks/{fr-BN5NKSAC.js → fr-MBMNQSQ6.js} +10 -9
  162. package/chunks/{gemini-2YFLVA2M.js → gemini-NRLXS5J7.js} +84 -70
  163. package/chunks/{geminiContentGenerator-XV6SV6CD.js → geminiContentGenerator-DLJPYYPA.js} +41 -5
  164. package/chunks/{glob-2H6CDDMK.js → glob-JPY3F74Y.js} +31 -25
  165. package/chunks/{grep-VXIUXM3C.js → grep-HW23YES5.js} +31 -25
  166. package/chunks/initializer-HWOZTC4A.js +67 -0
  167. package/chunks/{ja-7DEUB2TS.js → ja-HFV5QCHR.js} +10 -9
  168. package/chunks/{keychain-token-storage-YLMDLKXT.js → keychain-token-storage-GPAKRDKQ.js} +2 -2
  169. package/chunks/list-OZGAGRDG.js +71 -0
  170. package/chunks/loadedSettingsAdapter-RUTZVVIA.js +66 -0
  171. package/chunks/{loop-wakeup-CITGGQHA.js → loop-wakeup-G2EHBRGR.js} +5 -5
  172. package/chunks/{ls-R5WP4BGZ.js → ls-IDHUVSAG.js} +4 -4
  173. package/chunks/{lsp-GSICJMMW.js → lsp-FSRYXGFT.js} +2 -2
  174. package/chunks/{monitor-HKLRS2HQ.js → monitor-QVWEIXU3.js} +33 -26
  175. package/chunks/nonInteractiveCli-RV2WYPML.js +109 -0
  176. package/chunks/{notebook-edit-Y6SKSGTT.js → notebook-edit-B26W3FFH.js} +37 -34
  177. package/chunks/{openaiContentGenerator-J7G2FYJT.js → openaiContentGenerator-6TH7AZGQ.js} +15 -13
  178. package/chunks/pidfile-KW2QQJZI.js +73 -0
  179. package/chunks/{pt-TNZRNLQB.js → pt-2THT3YHA.js} +10 -9
  180. package/chunks/{qwenContentGenerator-5XIEM5GV.js → qwenContentGenerator-5IQAC7RN.js} +33 -27
  181. package/chunks/{qwenOAuth2-AG3PFFLZ.js → qwenOAuth2-LWLA6MQG.js} +5 -4
  182. package/chunks/{read-file-J7D552ZJ.js → read-file-CWFAHOI2.js} +11 -10
  183. package/chunks/{read-mcp-resource-DQQW7H7T.js → read-mcp-resource-PIJF24RO.js} +2 -2
  184. package/chunks/ripGrep-FTZGDKY4.js +60 -0
  185. package/chunks/{ru-NIAUBKCN.js → ru-Z67YJHBG.js} +10 -9
  186. package/chunks/{run-qwen-serve-SGEEVMXD.js → run-qwen-serve-FN2SW5ZI.js} +545 -97
  187. package/chunks/{scheduler-7CQI3S4Q.js → scheduler-BMS23O3I.js} +31 -25
  188. package/chunks/{send-message-EE4TWKVU.js → send-message-33PYFMCT.js} +20 -4
  189. package/chunks/serve-PXFYKL65.js +71 -0
  190. package/chunks/{server-4QMUWXPW.js → server-BRKALJ4F.js} +4132 -1175
  191. package/chunks/{session-NRO7UFOY.js → session-AQMISBP4.js} +205 -64
  192. package/chunks/{settings-SGWRDQSC.js → settings-JCPVUZMB.js} +35 -29
  193. package/chunks/{shell-62ANGIHL.js → shell-POHCRVN2.js} +31 -25
  194. package/chunks/{skill-VWAQAWW6.js → skill-QPEVHS6H.js} +13 -12
  195. package/chunks/spawnChannel-V3QT62TE.js +69 -0
  196. package/chunks/{src-KUV5J5Q2.js → src-IZSP7L6Y.js} +127 -32
  197. package/chunks/{startInteractiveUI-6EYJN3JV.js → startInteractiveUI-63KATISO.js} +16762 -16075
  198. package/chunks/{syntheticOutput-T5YHQ6S2.js → syntheticOutput-YOSWIHDJ.js} +3 -3
  199. package/chunks/{task-create-BLY7RHXX.js → task-create-NHS2NGB3.js} +9 -7
  200. package/chunks/{task-list-3QUYNVYR.js → task-list-5UIB564R.js} +7 -6
  201. package/chunks/{task-stop-U4RATK2V.js → task-stop-HGXZ2VZ3.js} +2 -2
  202. package/chunks/{task-update-U54O7KJX.js → task-update-IYYWUILG.js} +40 -7
  203. package/chunks/{team-create-LPOGCPM7.js → team-create-E5O2F3CA.js} +31 -25
  204. package/chunks/{team-delete-LKBYPASD.js → team-delete-FEKJCU4P.js} +7 -6
  205. package/chunks/team-plan-approval-2TU4YROB.js +202 -0
  206. package/chunks/theme-manager-37KOGPJL.js +64 -0
  207. package/chunks/{todoWrite-OYXKTVYN.js → todoWrite-TKMVZ2HT.js} +4 -4
  208. package/chunks/{tool-search-YPUFJ3ZQ.js → tool-search-4ZTGNSFI.js} +38 -11
  209. package/chunks/{trustedFolders-5B24TMEA.js → trustedFolders-RJ5WXJ6L.js} +32 -26
  210. package/chunks/{validateNonInterActiveAuth-BTDTYOVT.js → validateNonInterActiveAuth-ZWMHZAOK.js} +66 -55
  211. package/chunks/{web-fetch-EEB5LZYU.js → web-fetch-J4PIW36Q.js} +5 -4
  212. package/chunks/{workflow-LAWIW5LI.js → workflow-6PG2PC5M.js} +68 -39
  213. package/chunks/workspace-providers-status-4QIZN2W4.js +68 -0
  214. package/chunks/workspace-service-CYMCOEIO.js +78 -0
  215. package/chunks/{write-file-PCV226DJ.js → write-file-R354LGQF.js} +41 -40
  216. package/chunks/{zh-5EUYXASN.js → zh-SLZNSSZB.js} +32 -10
  217. package/chunks/{zh-TW-TR3XYGE5.js → zh-TW-P53C2SL5.js} +32 -10
  218. package/cli.js +6 -6
  219. package/locales/ca.js +15 -13
  220. package/locales/de.js +15 -13
  221. package/locales/en.js +52 -15
  222. package/locales/fr.js +15 -14
  223. package/locales/ja.js +15 -14
  224. package/locales/pt.js +15 -13
  225. package/locales/ru.js +15 -13
  226. package/locales/zh-TW.js +50 -14
  227. package/locales/zh.js +50 -14
  228. package/node_modules/@qwen-code/audio-capture/package.json +1 -1
  229. package/node_modules/@qwen-code/audio-capture/prebuilds/darwin-arm64/@qwen-code+audio-capture.node +0 -0
  230. package/node_modules/@qwen-code/audio-capture/prebuilds/darwin-x64/@qwen-code+audio-capture.node +0 -0
  231. package/node_modules/@qwen-code/audio-capture/prebuilds/win32-x64/@qwen-code+audio-capture.node +0 -0
  232. package/package.json +15 -6
  233. package/patches/chrome-devtools-mcp+1.4.0.patch +31 -0
  234. package/postinstall.js +50 -0
  235. package/web-shell/assets/{arc-DA7r8Ueo.js → arc-Bd3048nX.js} +1 -1
  236. package/web-shell/assets/{architectureDiagram-3BPJPVTR-ZDUxPtZ2.js → architectureDiagram-3BPJPVTR-DyzbzABx.js} +1 -1
  237. package/web-shell/assets/{blockDiagram-GPEHLZMM-vrtOf1kJ.js → blockDiagram-GPEHLZMM-CTjxm9Sy.js} +1 -1
  238. package/web-shell/assets/{c4Diagram-AAUBKEIU-BHFjzIi4.js → c4Diagram-AAUBKEIU-DSOFWNU6.js} +1 -1
  239. package/web-shell/assets/channel-oVoup_Ci.js +1 -0
  240. package/web-shell/assets/{chunk-2J33WTMH-mdeQVz59.js → chunk-2J33WTMH-D_PFdjo2.js} +1 -1
  241. package/web-shell/assets/{chunk-4BX2VUAB-BkKMlwBX.js → chunk-4BX2VUAB-52eG6wOT.js} +1 -1
  242. package/web-shell/assets/{chunk-55IACEB6-B7bnXNfu.js → chunk-55IACEB6-CGTZfKkC.js} +1 -1
  243. package/web-shell/assets/{chunk-727SXJPM-DGz8Mm6i.js → chunk-727SXJPM-CZ-qfRUm.js} +1 -1
  244. package/web-shell/assets/{chunk-AQP2D5EJ-BQAydUNk.js → chunk-AQP2D5EJ-B2dSDcqM.js} +1 -1
  245. package/web-shell/assets/{chunk-FMBD7UC4-GHr2LN_p.js → chunk-FMBD7UC4-Cfy3tyA0.js} +1 -1
  246. package/web-shell/assets/{chunk-ND2GUHAM-7R91lnUx.js → chunk-ND2GUHAM-CApNUpUq.js} +1 -1
  247. package/web-shell/assets/{chunk-QZHKN3VN-aTJD3lSQ.js → chunk-QZHKN3VN-CVBG8kgg.js} +1 -1
  248. package/web-shell/assets/classDiagram-4FO5ZUOK-B8WsaT2G.js +1 -0
  249. package/web-shell/assets/classDiagram-v2-Q7XG4LA2-B8WsaT2G.js +1 -0
  250. package/web-shell/assets/{cose-bilkent-S5V4N54A-BTf8t1RL.js → cose-bilkent-S5V4N54A-URd77-_-.js} +1 -1
  251. package/web-shell/assets/{dagre-BM42HDAG-_sgp3WuA.js → dagre-BM42HDAG-C07H4G9-.js} +1 -1
  252. package/web-shell/assets/{diagram-2AECGRRQ-CROtRy1w.js → diagram-2AECGRRQ-CYbpXrSn.js} +1 -1
  253. package/web-shell/assets/{diagram-5GNKFQAL-BqGRzwSc.js → diagram-5GNKFQAL-SI8s6fjX.js} +1 -1
  254. package/web-shell/assets/{diagram-KO2AKTUF-CMrcEtA7.js → diagram-KO2AKTUF-mBkC4RTW.js} +1 -1
  255. package/web-shell/assets/{diagram-LMA3HP47-BRydGNW5.js → diagram-LMA3HP47-CyBPDzWJ.js} +1 -1
  256. package/web-shell/assets/{diagram-OG6HWLK6-D0ilgNjV.js → diagram-OG6HWLK6-DwK7kBCu.js} +1 -1
  257. package/web-shell/assets/{erDiagram-TEJ5UH35-2lnQeJUM.js → erDiagram-TEJ5UH35-Ds1StEUd.js} +1 -1
  258. package/web-shell/assets/{flowDiagram-I6XJVG4X-C_ZMdQiC.js → flowDiagram-I6XJVG4X-BDC4-jHO.js} +1 -1
  259. package/web-shell/assets/{ganttDiagram-6RSMTGT7-Dw9V_Ny0.js → ganttDiagram-6RSMTGT7-CV2aHQqf.js} +1 -1
  260. package/web-shell/assets/{gitGraphDiagram-PVQCEYII-HDXI-F1B.js → gitGraphDiagram-PVQCEYII-B8RcuV_L.js} +1 -1
  261. package/web-shell/assets/index-B5WZ0qP1.js +717 -0
  262. package/web-shell/assets/index-CYE1IMd1.css +5 -0
  263. package/web-shell/assets/{infoDiagram-5YYISTIA-CuB8Ebua.js → infoDiagram-5YYISTIA-DUPoTf50.js} +1 -1
  264. package/web-shell/assets/{ishikawaDiagram-YF4QCWOH-D_tyhVwh.js → ishikawaDiagram-YF4QCWOH-FDYOx9rf.js} +1 -1
  265. package/web-shell/assets/{journeyDiagram-JHISSGLW-CgCwJQNU.js → journeyDiagram-JHISSGLW-C5_ohMRp.js} +1 -1
  266. package/web-shell/assets/{kanban-definition-UN3LZRKU-PZUtcbPx.js → kanban-definition-UN3LZRKU-Bn4UnOMG.js} +1 -1
  267. package/web-shell/assets/{linear-CZqRdJIU.js → linear-Bl0pKQtY.js} +1 -1
  268. package/web-shell/assets/{mermaid.core-CPW_4UEF.js → mermaid.core-sevQFoji.js} +72 -72
  269. package/web-shell/assets/{mindmap-definition-RKZ34NQL-D_Y75ID7.js → mindmap-definition-RKZ34NQL-DSH4JZF9.js} +1 -1
  270. package/web-shell/assets/{pieDiagram-4H26LBE5-Q4dTyLeF.js → pieDiagram-4H26LBE5-acWvdQTD.js} +1 -1
  271. package/web-shell/assets/{quadrantDiagram-W4KKPZXB-ckw329Ft.js → quadrantDiagram-W4KKPZXB-DlggX4Sm.js} +1 -1
  272. package/web-shell/assets/{requirementDiagram-4Y6WPE33-DbVDm4It.js → requirementDiagram-4Y6WPE33--mMqwmV8.js} +1 -1
  273. package/web-shell/assets/{sankeyDiagram-5OEKKPKP-CHuZ_ecH.js → sankeyDiagram-5OEKKPKP-BtV9tZ77.js} +1 -1
  274. package/web-shell/assets/{sequenceDiagram-3UESZ5HK-DU2YSpq-.js → sequenceDiagram-3UESZ5HK-iDQzPFlj.js} +1 -1
  275. package/web-shell/assets/{stateDiagram-AJRCARHV-CA9PEBUo.js → stateDiagram-AJRCARHV-CsCyQLZ3.js} +1 -1
  276. package/web-shell/assets/stateDiagram-v2-BHNVJYJU-CPapECG9.js +1 -0
  277. package/web-shell/assets/{timeline-definition-PNZ67QCA-Bhlo4WVc.js → timeline-definition-PNZ67QCA-Dl0rnKvr.js} +1 -1
  278. package/web-shell/assets/{vennDiagram-CIIHVFJN-6xtPLsyl.js → vennDiagram-CIIHVFJN-qF7nTZzA.js} +1 -1
  279. package/web-shell/assets/{wardley-L42UT6IY-3eozmXPf.js → wardley-L42UT6IY-fWoK9i7F.js} +1 -1
  280. package/web-shell/assets/{wardleyDiagram-YWT4CUSO-BOn6hKss.js → wardleyDiagram-YWT4CUSO-pHGfkMnc.js} +1 -1
  281. package/web-shell/assets/{xychartDiagram-2RQKCTM6-BsP4-I_y.js → xychartDiagram-2RQKCTM6-3Doc4Trz.js} +1 -1
  282. package/web-shell/index.html +18 -2
  283. package/chunks/MaxSizedBox-QNF3EIGU.js +0 -63
  284. package/chunks/agent-headless-RQ5BTX3G.js +0 -56
  285. package/chunks/bridge-JL2PMCIW.js +0 -66
  286. package/chunks/chunk-5GOVVXEV.js +0 -73
  287. package/chunks/chunk-BC3C4F3Q.js +0 -40
  288. package/chunks/chunk-BW4KCQFD.js +0 -1103
  289. package/chunks/chunk-RKIZXJMB.js +0 -13
  290. package/chunks/contextCommand-WYU6OQ2F.js +0 -61
  291. package/chunks/daemon-status-provider-HH2EC5FI.js +0 -65
  292. package/chunks/headlessSafetyWarnings-TBE35GJ4.js +0 -55
  293. package/chunks/initializer-2Y7SLOKV.js +0 -61
  294. package/chunks/list-ZAFHZ7OS.js +0 -65
  295. package/chunks/loadedSettingsAdapter-6KJXD6M3.js +0 -60
  296. package/chunks/nonInteractiveCli-5W4GIZQM.js +0 -96
  297. package/chunks/ripGrep-IEIFIWP4.js +0 -54
  298. package/chunks/serve-V6GVZR6U.js +0 -65
  299. package/chunks/spawnChannel-3VHW5EPY.js +0 -63
  300. package/chunks/theme-manager-LYLGAERF.js +0 -58
  301. package/chunks/workspace-providers-status-PIAFVYHA.js +0 -62
  302. package/chunks/workspace-service-YAIOD4HG.js +0 -71
  303. package/web-shell/assets/channel-iRo68q4Q.js +0 -1
  304. package/web-shell/assets/classDiagram-4FO5ZUOK-iHF24a8X.js +0 -1
  305. package/web-shell/assets/classDiagram-v2-Q7XG4LA2-iHF24a8X.js +0 -1
  306. package/web-shell/assets/index-CHmzAWys.css +0 -5
  307. package/web-shell/assets/index-w6GTOHh1.js +0 -713
  308. package/web-shell/assets/stateDiagram-v2-BHNVJYJU-t1LRI0oY.js +0 -1
@@ -91,6 +91,9 @@ Settings are organized into categories. Most settings should be placed within th
91
91
  | `general.language` | enum | Language for the user interface. Use `"auto"` to detect from system settings, or a language code (e.g. `"zh-CN"`, `"fr"`). Custom codes can be added by placing JS locale files in `~/.qwen/locales/`. See [i18n](../features/language). Requires restart. | `"auto"` |
92
92
  | `general.outputLanguage` | string | Language for model output. Use `"auto"` to detect from system settings, or set a specific language. Requires restart. | `"auto"` |
93
93
  | `general.dynamicCommandTranslation` | boolean | Enable AI translation of dynamic slash-command descriptions. When disabled, dynamic commands keep their original descriptions and skip translation model calls. | `false` |
94
+ | `general.terminalBell` | boolean | Play a terminal bell sound when a response completes or needs approval. | `true` |
95
+ | `general.preventSystemSleep` | boolean | Prevent the system from sleeping while Qwen Code is streaming a model response or executing tools. Idle prompt time and permission prompts do not inhibit sleep. Read once at startup, so changes take effect after restart. | `true` |
96
+ | `general.chatRecording` | boolean | Save chat history to disk. Disabling this also prevents `--continue` and `--resume` from working. Requires restart. | `true` |
94
97
 
95
98
  #### output
96
99
 
@@ -116,6 +119,7 @@ Settings are organized into categories. Most settings should be placed within th
116
119
  | `ui.renderMode` | string | Default Markdown display mode. Use `"render"` for rich visual previews or `"raw"` to show source-oriented Markdown by default. Toggle during a session with `Alt/Option+M`; on macOS the terminal must send Option as Meta. See [Markdown Rendering](../features/markdown-rendering). | `"render"` |
117
120
  | `ui.showCitations` | boolean | Show citations for generated text in the chat. | `false` |
118
121
  | `ui.history.collapseOnResume` | boolean | Whether to collapse history by default when resuming a session. Can be toggled via `/history collapse-on-resume` and `/history expand-on-resume`. | `false` |
122
+ | `ui.history.collapsePreviewCount` | number | Number of most recent user turns to keep visible when `ui.history.collapseOnResume` is enabled. `0` collapses all restored history by default; `-1` shows all restored history. | `0` |
119
123
  | `ui.compactMode` | boolean | Hide tool output and thinking for a cleaner view. Toggle with `Ctrl+O` during a session or via the Settings dialog. Tool approval prompts are never hidden, even in compact mode. The setting persists across sessions. | `false` |
120
124
  | `ui.shellOutputMaxLines` | number | Max number of shell output lines shown inline. Set to `0` to disable the cap and show full output. Hidden lines are surfaced via the `+N lines` indicator. Errors, `!`-prefix user-initiated commands, confirming tools, and focused embedded shells always show full output. | `5` |
121
125
  | `ui.enableWelcomeBack` | boolean | Show welcome back dialog when returning to a project with conversation history. When enabled, Qwen Code will automatically detect if you're returning to a project with a previously generated project summary (`.qwen/PROJECT_SUMMARY.md`) and show a dialog allowing you to continue your previous conversation or start fresh. If you choose **Start new chat session**, that choice is remembered for the current project until the project summary changes. This feature integrates with the `/summary` command and quit confirmation dialog. | `true` |
@@ -126,6 +130,12 @@ Settings are organized into categories. Most settings should be placed within th
126
130
  | `ui.enableFollowupSuggestions` | boolean | Enable [followup suggestions](../features/followup-suggestions) that predict what you want to type next after the model responds. Suggestions appear as placeholder text and are accepted with Tab, Enter, or Right Arrow (which fill the input — they do not auto-submit). On by default; set to `false` to opt out. | `true` |
127
131
  | `ui.enableCacheSharing` | boolean | Use cache-aware forked queries for suggestion generation. Reduces cost on providers that support prefix caching (experimental). | `true` |
128
132
  | `ui.enableSpeculation` | boolean | Speculatively execute accepted suggestions before submission. Results appear instantly when you accept (experimental). | `false` |
133
+ | `ui.showStatusInTitle` | boolean | Show the Qwen Code session name and status in the terminal window title. | `true` |
134
+ | `ui.disableWorkflowKeywordTrigger` | boolean | When `true`, mentioning the word `workflow` in a prompt no longer softly steers the turn toward the Workflow tool (and the Footer `workflow active` indicator is suppressed). Only applies when workflows are enabled. | `false` |
135
+ | `ui.enableUserFeedback` | boolean | Show an optional feedback dialog after conversations to help improve Qwen performance. | `true` |
136
+ | `ui.compactInline` | boolean | Compact tool display within each group instead of merging across groups. Requires `ui.compactMode` to be enabled. Requires restart. | `false` |
137
+ | `ui.useTerminalBuffer` | boolean | Render conversation history in an in-app scrollable viewport instead of the terminal scrollback buffer. Recommended if you see flicker, scroll-storm, or interface freeze on long sessions. Scroll with `Shift+↑/↓` (line), `PgUp`/`PgDn` (page), `Ctrl+Home/End` (top/bottom), or the mouse wheel. Does not use the host terminal scrollback while enabled; hold `Shift` (or `Option` on macOS) while dragging for native text selection. | `false` |
138
+ | `ui.hideBuiltinWorktreeIndicator` | boolean | Hide the built-in `⎇ worktree-<branch> (<slug>)` line in the Footer. The worktree state is still passed to custom statusline scripts via the stdin payload. Keep at the default unless your custom statusline renders the worktree itself. | `false` |
129
139
 
130
140
  #### ide
131
141
 
@@ -145,11 +155,13 @@ Settings are organized into categories. Most settings should be placed within th
145
155
  | Setting | Type | Description | Default |
146
156
  | -------------------------------------------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- |
147
157
  | `model.name` | string | The Qwen model to use for conversations. | `undefined` |
158
+ | `model.baseUrl` | string | Persisted automatically by the model picker to disambiguate when multiple `modelProviders` entries share the same model id. Not intended to be set by hand — use the `/model` picker or a `modelProviders` entry instead; a stale hand-edited value can silently route requests to a different same-id provider. | `undefined` |
159
+ | `model.sessionTokenLimit` | number | Maximum recorded prompt token count allowed before sending the next message. `-1` means unlimited; `0` is also treated as unlimited (unlike `model.maxToolCalls`, where `0` disallows all calls). When the recorded prompt count exceeds the limit, the next send is dropped (the session is not aborted). | `-1` |
148
160
  | `model.maxSessionTurns` | number | Maximum number of user/model/tool turns to keep in a session. -1 means unlimited. | `-1` |
149
161
  | `model.maxWallTimeSeconds` | number | Wall-clock budget for headless / unattended runs, in seconds. `-1` means unlimited. Overridable per-invocation via `--max-wall-time`, which requires a positive duration (`90`, `30s`, `5m`, `1h`, `1.5h`); the minimum is 1 second — sub-second values (`500ms`, `0.5`) are rejected as typos. Omit the flag to fall back to this setting. Aborts with exit code 55 when exceeded. | `-1` |
150
162
  | `model.maxToolCalls` | number | Cumulative tool-call budget for a run (counts every executed tool, success or failure; `structured_output` under `--json-schema` is exempt). `-1` means unlimited; `0` means "no tool calls allowed". Capped at 1,000,000 to catch typos. Overridable via `--max-tool-calls`. Aborts with exit code 55 when exceeded. | `-1` |
151
163
  | `model.generationConfig` | object | Advanced overrides passed to the underlying content generator. Supports request controls such as `timeout`, `maxRetries`, `enableCacheControl`, `splitToolMedia` (default `true`; splits tool-returned media — including images read by the built-in read_file — into a follow-up user message instead of the spec-violating `role: "tool"` message, so strict OpenAI-compatible servers like doubao / new-api / LM Studio can see it; set `false` to restore the legacy embed-in-tool behavior), `toolResultContentFormat` (default `"parts"`; set `"string"` only for legacy OpenAI-compatible runtimes whose tool templates ignore text content parts), `contextWindowSize` (override model's context window size), `modalities` (override auto-detected input modalities), `customHeaders` (custom HTTP headers for API requests), and `extra_body` (additional body parameters for OpenAI-compatible API requests only), along with fine-tuning knobs under `samplingParams` (for example `temperature`, `top_p`, `max_tokens`). Leave unset to rely on provider defaults. | `undefined` |
152
- | `model.chatCompression.contextPercentageThreshold` | number | **REMOVED.** Auto-compaction now uses a three-tier threshold ladder (warn / auto / hard) computed internally from the model's context window via the `computeThresholds()` function — no longer user-configurable. Setting this field in `settings.json` is silently ignored (no startup warning). There is currently no replacement for "disable compression entirely" — reactive overflow recovery remains the safety net at the API layer if compression itself fails. (See PR #4345 / `docs/design/auto-compaction-threshold-redesign.md` for the redesign rationale.) | `N/A` |
164
+ | `model.chatCompression.contextPercentageThreshold` | number | **REMOVED.** Replaced by `context.autoCompactThreshold` (see `#### context` section below). Auto-compaction now uses a three-tier threshold ladder (warn / auto / hard) computed internally from the model's context window via the `computeThresholds()` function. The old setting is silently ignored (no startup warning). See PR #4345 / `docs/design/auto-compaction-threshold-redesign.md` for the redesign rationale. | `N/A` |
153
165
  | `model.chatCompression.maxRecentFilesToRetain` | number | Number of most-recently-touched files whose current content is restored (embedded if small, otherwise referenced by path) into history after auto-compaction. `0` restores none. Env override: `QWEN_COMPACT_MAX_RECENT_FILES`. | `5` |
154
166
  | `model.chatCompression.maxRecentImagesToRetain` | number | Number of most-recent images (tool screenshots / user pastes) restored into history after auto-compaction. `0` restores none. Env override: `QWEN_COMPACT_MAX_RECENT_IMAGES`. | `3` |
155
167
  | `model.chatCompression.enableScreenshotTrigger` | boolean | When `true`, auto-compaction also fires once the number of tool-returned images accumulated in history reaches `screenshotTriggerThreshold`, independent of token usage — aimed at computer-use sessions where frequent screenshots dilute model attention. Counts only images returned inside tool results, not user-pasted images. Env override: `QWEN_COMPACT_SCREENSHOT_TRIGGER` (`1`/`true`/`0`/`false`). | `true` |
@@ -232,22 +244,35 @@ The `extra_body` field allows you to add custom parameters to the request body s
232
244
  | ----------- | ------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
233
245
  | `fastModel` | string | Model used for generating [prompt suggestions](../features/followup-suggestions) and speculative execution. Leave empty to use the main model. A smaller/faster model (e.g., `qwen3-coder-flash`) reduces latency and cost. Can also be set via `/model --fast`. | `""` |
234
246
 
247
+ #### visionModel
248
+
249
+ | Setting | Type | Description | Default |
250
+ | ------------- | ------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
251
+ | `visionModel` | string | Image-capable model used as the vision bridge: when a text-only main model receives an image, it is transcribed by this model first. Leave empty to auto-pick a same-provider vision model. Can also be set via `/model --vision`. | `""` |
252
+
253
+ #### voiceModel
254
+
255
+ | Setting | Type | Description | Default |
256
+ | ------------ | ------ | ------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
257
+ | `voiceModel` | string | Model used for voice transcription. Leave empty to keep voice dictation disabled until a voice model is selected. Can also be set via `/model --voice`. | `""` |
258
+
235
259
  #### context
236
260
 
237
- | Setting | Type | Description | Default |
238
- | ----------------------------------------------------------- | -------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------- |
239
- | `context.fileName` | string or array of strings | The name of the context file(s). | `undefined` |
240
- | `context.importFormat` | string | The format to use when importing memory. | `undefined` |
241
- | `context.includeDirectories` | array | Additional directories to include in the workspace context. Specifies an array of additional absolute or relative paths to include in the workspace context. Missing directories will be skipped with a warning by default. Paths can use `~` to refer to the user's home directory. This setting can be combined with the `--include-directories` command-line flag. | `[]` |
242
- | `context.loadFromIncludeDirectories` | boolean | Controls the behavior of the `/memory refresh` command. If set to `true`, `QWEN.md` files should be loaded from all directories that are added. If set to `false`, `QWEN.md` should only be loaded from the current directory. | `false` |
243
- | `context.fileFiltering.respectGitIgnore` | boolean | Respect .gitignore files when searching. | `true` |
244
- | `context.fileFiltering.respectQwenIgnore` | boolean | Respect .qwenignore and configured custom ignore files when searching. | `true` |
245
- | `context.fileFiltering.customIgnoreFiles` | array | Project-root-relative ignore files to use instead of the default compatibility files (`.agentignore`, `.aiignore`) when `respectQwenIgnore` is enabled. `.qwenignore` is always included. | `[".agentignore", ".aiignore"]` |
246
- | `context.fileFiltering.enableRecursiveFileSearch` | boolean | Whether to enable searching recursively for filenames under the current tree when completing `@` prefixes in the prompt. | `true` |
247
- | `context.fileFiltering.enableFuzzySearch` | boolean | When `true`, enables fuzzy search capabilities when searching for files. Set to `false` to improve performance on projects with a large number of files. | `true` |
248
- | `context.clearContextOnIdle.toolResultsThresholdMinutes` | number | Minutes of inactivity before clearing old tool result content. Use `-1` to disable the idle trigger. | `60` |
249
- | `context.clearContextOnIdle.toolResultsNumToKeep` | integer | Integer number of most-recent compactable tool results to preserve when clearing. Values below 1 are floored to 1. | `5` |
250
- | `context.clearContextOnIdle.toolResultsTotalCharsThreshold` | number | Total compactable tool result output characters allowed in history before clearing oldest results. Use `-1` to disable the size trigger. This is a soft threshold: protected recent tool results may keep the total above it. | `500000` |
261
+ | Setting | Type | Description | Default |
262
+ | ----------------------------------------------------------- | -------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------- |
263
+ | `context.fileName` | string or array of strings | The name of the context file(s). | `undefined` |
264
+ | `context.autoCompactThreshold` | number | Fraction of the context window at which auto-compaction triggers. Must be greater than 0 and at most 1. Default is `0.7` (70%). For large context windows (>110K tokens), the absolute branch of the three-tier threshold system dominates, so values below ~0.7 may have no visible effect. Custom thresholds primarily affect small-window models (≤128K). Replaces the old `model.chatCompression.contextPercentageThreshold`. | `undefined` (uses internal 0.7) |
265
+ | `context.importFormat` | string | The format to use when importing memory. | `undefined` |
266
+ | `context.includeDirectories` | array | Additional directories to include in the workspace context. Specifies an array of additional absolute or relative paths to include in the workspace context. Missing directories will be skipped with a warning by default. Paths can use `~` to refer to the user's home directory. This setting can be combined with the `--include-directories` command-line flag. | `[]` |
267
+ | `context.loadFromIncludeDirectories` | boolean | Controls the behavior of the `/memory refresh` command. If set to `true`, `QWEN.md` files should be loaded from all directories that are added. If set to `false`, `QWEN.md` should only be loaded from the current directory. | `false` |
268
+ | `context.fileFiltering.respectGitIgnore` | boolean | Respect .gitignore files when searching. | `true` |
269
+ | `context.fileFiltering.respectQwenIgnore` | boolean | Respect .qwenignore and configured custom ignore files when searching. | `true` |
270
+ | `context.fileFiltering.customIgnoreFiles` | array | Project-root-relative ignore files to use instead of the default compatibility files (`.agentignore`, `.aiignore`) when `respectQwenIgnore` is enabled. `.qwenignore` is always included. | `[".agentignore", ".aiignore"]` |
271
+ | `context.fileFiltering.enableRecursiveFileSearch` | boolean | Whether to enable searching recursively for filenames under the current tree when completing `@` prefixes in the prompt. | `true` |
272
+ | `context.fileFiltering.enableFuzzySearch` | boolean | When `true`, enables fuzzy search capabilities when searching for files. Set to `false` to improve performance on projects with a large number of files. | `true` |
273
+ | `context.clearContextOnIdle.toolResultsThresholdMinutes` | number | Minutes of inactivity before clearing old tool result content. Use `-1` to disable the idle trigger. | `60` |
274
+ | `context.clearContextOnIdle.toolResultsNumToKeep` | integer | Integer number of most-recent compactable tool results to preserve when clearing. Values below 1 are floored to 1. | `5` |
275
+ | `context.clearContextOnIdle.toolResultsTotalCharsThreshold` | number | Total compactable tool result output characters allowed in history before clearing oldest results. Use `-1` to disable the size trigger. This is a soft threshold: protected recent tool results may keep the total above it. | `500000` |
251
276
 
252
277
  #### Troubleshooting File Search Performance
253
278
 
@@ -276,6 +301,7 @@ If you are experiencing performance issues with file searching (e.g., with `@` c
276
301
  | `tools.truncateToolOutputLines` | number | Maximum lines or entries kept when truncating tool output. Applies to Shell, Grep, Glob, ReadFile and ReadManyFiles tools. | `1000` | Requires restart: Yes |
277
302
  | `tools.computerUse.enabled` | boolean | Enable the built-in Computer Use tools (cua-driver native desktop automation). When `true` (default), the `computer_use__*` tools are registered as deferred built-ins; the first invocation downloads the pinned, signed cua-driver binary into `~/.qwen/computer-use/` and walks through macOS Accessibility / Screen Recording permissions. | `true` | Requires restart: Yes |
278
303
  | `tools.computerUse.maxImageDimension` | number | Longest-edge pixel cap applied to cua-driver screenshots (via `set_config`'s `max_image_dimension`). `-1` (default) keeps cua-driver's built-in default (1568); `0` disables resizing (full resolution); a positive value caps the longest edge. Lower caps cut vision-token cost at the expense of fine detail. | `-1` | Requires restart: Yes. Env override: `QWEN_COMPUTER_USE_MAX_IMAGE_DIMENSION` (a non-negative integer; takes precedence over this setting) |
304
+ | `tools.toolSearch.enabled` | boolean | Load MCP tools on demand via ToolSearch to reduce prompt size. Disable this for models that rely on prefix-based KV caching (e.g. DeepSeek) to keep the prompt prefix stable and maximize cache hit rates. | `true` | Requires restart: Yes |
279
305
 
280
306
  > [!note]
281
307
  >
@@ -283,12 +309,14 @@ If you are experiencing performance issues with file searching (e.g., with `@` c
283
309
 
284
310
  #### memory
285
311
 
286
- | Setting | Type | Description | Default |
287
- | -------------------------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------ | ------- |
288
- | `memory.enableManagedAutoMemory` | boolean | Enable background extraction of memories from conversations. | `true` |
289
- | `memory.enableManagedAutoDream` | boolean | Enable automatic consolidation (deduplication and cleanup) of collected memories. | `true` |
290
- | `memory.enableAutoSkill` | boolean | Enable background review for reusable project skills after tool-heavy sessions. | `true` |
291
- | `memory.autoSkillConfirm` | boolean | Ask for confirmation before auto-generated skills are added to the skill library. When off, auto-skills are saved immediately. | `true` |
312
+ | Setting | Type | Description | Default |
313
+ | -------------------------------- | ------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
314
+ | `memory.enableManagedAutoMemory` | boolean | Enable background extraction of memories from conversations. | `true` |
315
+ | `memory.enableManagedAutoDream` | boolean | Enable automatic consolidation (deduplication and cleanup) of collected memories. | `true` |
316
+ | `memory.enableAutoSkill` | boolean | Enable background review for reusable project skills after tool-heavy sessions. | `true` |
317
+ | `memory.autoSkillConfirm` | boolean | Ask for confirmation before auto-generated skills are added to the skill library. When off, auto-skills are saved immediately. | `true` |
318
+ | `memory.enableTeamMemory` | boolean | Enable a project memory tier shared with collaborators via the git-tracked `.qwen/team-memory/` directory. Writes to it are secret-scanned and reviewable in the git diff. | `false` |
319
+ | `memory.enableTeamMemorySync` | boolean | When team memory is enabled, automatically commit, fast-forward-pull, and push the `.qwen/team-memory/` directory at session start so collaborators stay in sync. Requires a configured git upstream. | `false` |
292
320
 
293
321
  See [Memory](../features/memory) for details on how auto-memory works and how to use the `/memory`, `/remember`, and `/dream` commands.
294
322
 
@@ -422,11 +450,11 @@ execute when typed.
422
450
 
423
451
  #### mcp
424
452
 
425
- | Setting | Type | Description | Default |
426
- | ------------------- | ---------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- |
427
- | `mcp.serverCommand` | string | Command to start an MCP server. | `undefined` |
428
- | `mcp.allowed` | array of strings | An allowlist of MCP servers to allow. Allows you to specify a list of MCP server names that should be made available to the model. This can be used to restrict the set of MCP servers to connect to. Note that this will be ignored if `--allowed-mcp-server-names` is set. | `undefined` |
429
- | `mcp.excluded` | array of strings | A denylist of MCP servers to exclude. A server listed in both `mcp.excluded` and `mcp.allowed` is excluded. Note that this will be ignored if `--allowed-mcp-server-names` is set. | `undefined` |
453
+ | Setting | Type | Description | Default |
454
+ | ------------------- | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- |
455
+ | `mcp.serverCommand` | string | Command to start an MCP server. | `undefined` |
456
+ | `mcp.allowed` | array of strings | An allowlist of MCP servers to allow. Allows you to specify a list of MCP server names that should be made available to the model. This can be used to restrict the set of MCP servers to connect to. Supports glob patterns (`*` matches any sequence, `?` matches a single character — e.g. `"*puppeteer*"`); entries without glob characters are matched exactly. Note that this will be ignored if `--allowed-mcp-server-names` is set. | `undefined` |
457
+ | `mcp.excluded` | array of strings | A denylist of MCP servers to exclude. A server listed in both `mcp.excluded` and `mcp.allowed` is excluded. Supports glob patterns (`*`, `?`) the same way as `mcp.allowed`. Note that this will be ignored if `--allowed-mcp-server-names` is set. | `undefined` |
430
458
 
431
459
  > [!note]
432
460
  >
@@ -443,12 +471,14 @@ LSP server configuration is done through `.lsp.json` files in your project root
443
471
 
444
472
  #### security
445
473
 
446
- | Setting | Type | Description | Default |
447
- | ------------------------------ | ------- | ------------------------------------------------- | ----------- |
448
- | `security.folderTrust.enabled` | boolean | Setting to track whether Folder trust is enabled. | `false` |
449
- | `security.auth.selectedType` | string | The currently selected authentication type. | `undefined` |
450
- | `security.auth.enforcedType` | string | The required auth type (useful for enterprises). | `undefined` |
451
- | `security.auth.useExternal` | boolean | Whether to use an external authentication flow. | `undefined` |
474
+ | Setting | Type | Description | Default |
475
+ | ------------------------------ | ------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------- |
476
+ | `security.folderTrust.enabled` | boolean | Setting to track whether Folder trust is enabled. | `false` |
477
+ | `security.auth.selectedType` | string | The currently selected authentication type. | `undefined` |
478
+ | `security.auth.enforcedType` | string | The required auth type (useful for enterprises). | `undefined` |
479
+ | `security.auth.useExternal` | boolean | Whether to use an external authentication flow. | `undefined` |
480
+ | `security.auth.apiKey` | string | **Deprecated.** API key for OpenAI-compatible authentication. Migrate to `modelProviders` with `envKey` instead — see [Model Providers](./model-providers). | `undefined` |
481
+ | `security.auth.baseUrl` | string | **Deprecated.** Base URL for the OpenAI-compatible API. Migrate to `modelProviders` instead — see [Model Providers](./model-providers). | `undefined` |
452
482
 
453
483
  #### advanced
454
484
 
@@ -300,6 +300,9 @@ reason inline and decide whether to switch to Ask Permissions Mode for that step
300
300
  "deny": ["Any network call to intranet.example.com"],
301
301
  },
302
302
  "environment": ["Open-source monorepo; commits are signed"],
303
+ // Optional: route ALL shell commands (including read-only ones like
304
+ // ls, cat) through the classifier for defense-in-depth.
305
+ // "classifyAllShell": true,
303
306
  },
304
307
  },
305
308
  }
@@ -43,6 +43,14 @@ The classifier uses your configured fast model
43
43
  (`/model --fast`). If no fast model is configured, the main session
44
44
  model is used instead.
45
45
 
46
+ > [!tip]
47
+ >
48
+ > Shell commands that the permission system detects as read-only (e.g.
49
+ > `ls`, `cat`, `git log`) are auto-approved before reaching the
50
+ > classifier. Set `permissions.autoMode.classifyAllShell: true` to
51
+ > override this and route all shell commands through the classifier —
52
+ > see [Classify all shell commands](#classify-all-shell-commands) below.
53
+
46
54
  ## Hard rules still win
47
55
 
48
56
  Auto Mode does **not** replace hard permission rules. Before the classifier
@@ -145,6 +153,38 @@ To keep the classifier system prompt small:
145
153
  way other permission settings are: arrays are concatenated and
146
154
  de-duplicated.
147
155
 
156
+ ### Classify all shell commands
157
+
158
+ By default, read-only shell commands (`ls`, `cat`, `git status`, …) are
159
+ auto-approved without invoking the classifier — the permission system
160
+ detects them as safe at layer 3 and skips the classifier entirely. Set
161
+ `classifyAllShell` to `true` to force **every** shell command through
162
+ the classifier, including read-only ones:
163
+
164
+ ```json
165
+ {
166
+ "permissions": {
167
+ "autoMode": {
168
+ "classifyAllShell": true
169
+ }
170
+ }
171
+ }
172
+ ```
173
+
174
+ This is useful for production or high-security environments where you
175
+ want defense-in-depth: even seemingly harmless commands are reviewed by
176
+ the classifier before execution. The trade-off is added latency (~300ms
177
+ per read-only shell call) and reliance on classifier availability — if
178
+ the classifier API is unreachable, read-only shell commands will also be
179
+ blocked (fail-closed).
180
+
181
+ > [!note]
182
+ >
183
+ > `classifyAllShell` only affects shell commands (`run_shell_command` and
184
+ > `monitor`). Built-in read-only tools (`read_file`, `grep_search`,
185
+ > `glob`, `list_directory`, etc.) are unaffected and still use the
186
+ > fast-path allowlist.
187
+
148
188
  ## Reading the decision
149
189
 
150
190
  When the classifier blocks an action, the tool call fails with one of
@@ -47,24 +47,25 @@ Channels are configured under the `channels` key in `settings.json`. Each channe
47
47
 
48
48
  ### Options
49
49
 
50
- | Option | Required | Description |
51
- | ------------------------ | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------- |
52
- | `type` | Yes | Channel type: `telegram`, `weixin`, `qq`, `dingtalk`, `feishu`, or a custom type from an extension (see [Plugins](./plugins)) |
53
- | `token` | Telegram | Bot token. Supports `$ENV_VAR` syntax to read from environment variables. Not needed for WeChat or DingTalk |
54
- | `clientId` | DingTalk | DingTalk AppKey. Supports `$ENV_VAR` syntax |
55
- | `clientSecret` | DingTalk | DingTalk AppSecret. Supports `$ENV_VAR` syntax |
56
- | `model` | No | Model to use for this channel (e.g., `qwen3.5-plus`). Overrides the default model. Useful for multimodal models that support image input |
57
- | `senderPolicy` | No | Who can talk to the bot: `allowlist` (default), `open`, or `pairing` |
58
- | `allowedUsers` | No | List of user IDs allowed to use the bot (used by `allowlist` and `pairing` policies) |
59
- | `sessionScope` | No | How sessions are scoped: `user` (default), `thread`, or `single` |
60
- | `cwd` | No | Working directory for the agent. Defaults to the current directory |
61
- | `instructions` | No | Custom instructions prepended to the first message of each session |
62
- | `groupPolicy` | No | Group chat access: `disabled` (default), `allowlist`, or `open`. See [Group Chats](#group-chats) |
63
- | `groups` | No | Per-group settings. Keys are group chat IDs or `"*"` for defaults. See [Group Chats](#group-chats) |
64
- | `dispatchMode` | No | What happens when you send a message while the bot is busy: `steer` (default), `collect`, or `followup`. See [Dispatch Modes](#dispatch-modes) |
65
- | `blockStreaming` | No | Progressive response delivery: `on` or `off` (default). See [Block Streaming](#block-streaming) |
66
- | `blockStreamingChunk` | No | Chunk size bounds: `{ "minChars": 400, "maxChars": 1000 }`. See [Block Streaming](#block-streaming) |
67
- | `blockStreamingCoalesce` | No | Idle flush: `{ "idleMs": 1500 }`. See [Block Streaming](#block-streaming) |
50
+ | Option | Required | Description |
51
+ | ------------------------ | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
52
+ | `type` | Yes | Channel type: `telegram`, `weixin`, `qq`, `dingtalk`, `feishu`, or a custom type from an extension (see [Plugins](./plugins)) |
53
+ | `token` | Telegram | Bot token. Supports `$ENV_VAR` syntax to read from environment variables. Not needed for WeChat or DingTalk |
54
+ | `clientId` | DingTalk | DingTalk AppKey. Supports `$ENV_VAR` syntax |
55
+ | `clientSecret` | DingTalk | DingTalk AppSecret. Supports `$ENV_VAR` syntax |
56
+ | `model` | No | Model to use for this channel (e.g., `qwen3.5-plus`). Overrides the default model. Useful for multimodal models that support image input |
57
+ | `senderPolicy` | No | Who can talk to the bot: `allowlist` (default), `open`, or `pairing` |
58
+ | `allowedUsers` | No | List of user IDs allowed to use the bot (used by `allowlist` and `pairing` policies) |
59
+ | `sessionScope` | No | How sessions are scoped: `user` (default), `thread`, or `single` |
60
+ | `cwd` | No | Working directory for the agent. Defaults to the current directory |
61
+ | `instructions` | No | Custom instructions prepended to the first message of each session |
62
+ | `groupPolicy` | No | Group chat access: `disabled` (default), `allowlist`, or `open`. See [Group Chats](#group-chats) |
63
+ | `groupHistoryLimit` | No | Opt-in group history backfill. `0` or omitted disables it. A positive number persists that many authorized, unmentioned group messages for the next bot mention/reply. |
64
+ | `groups` | No | Per-group settings. Keys are group chat IDs or `"*"` for defaults. See [Group Chats](#group-chats) |
65
+ | `dispatchMode` | No | What happens when you send a message while the bot is busy: `steer` (default), `collect`, or `followup`. See [Dispatch Modes](#dispatch-modes) |
66
+ | `blockStreaming` | No | Progressive response delivery: `on` or `off` (default). See [Block Streaming](#block-streaming) |
67
+ | `blockStreamingChunk` | No | Chunk size bounds: `{ "minChars": 400, "maxChars": 1000 }`. See [Block Streaming](#block-streaming) |
68
+ | `blockStreamingCoalesce` | No | Idle flush: `{ "idleMs": 1500 }`. See [Block Streaming](#block-streaming) |
68
69
 
69
70
  ### Sender Policy
70
71
 
@@ -82,6 +83,18 @@ Controls how conversation sessions are managed:
82
83
  - **`thread`** — One session per thread/topic. Useful for group chats with threads.
83
84
  - **`single`** — One shared session for all users. Everyone shares the same conversation.
84
85
 
86
+ ### Channel Memory
87
+
88
+ Channel memory lets an authorized channel member save stable context for one chat or thread. Qwen Code injects that memory when a fresh channel session starts, including after `/clear`.
89
+
90
+ Commands:
91
+
92
+ - `/remember-channel <text>` saves a memory line for the current chat or thread.
93
+ - `/channel-memory` shows saved memory for the current chat or thread.
94
+ - `/forget-channel confirm` clears saved memory for the current chat or thread.
95
+
96
+ Only users listed in `allowedUsers` can read, write, or clear channel memory. If `allowedUsers` is empty, channel memory commands are disabled for everyone.
97
+
85
98
  ### Token Security
86
99
 
87
100
  Bot tokens should not be stored directly in `settings.json`. Instead, use environment variable references:
@@ -158,6 +171,38 @@ Configure per-group with the `groups` setting:
158
171
  - **Group chat ID** — Override settings for a specific group. Overrides `"*"` defaults.
159
172
  - **`requireMention`** (default: `true`) — When `true`, the bot only responds to messages that @mention it or reply to one of its messages. When `false`, the bot responds to all messages (useful for dedicated task groups).
160
173
 
174
+ ### Group History Backfill
175
+
176
+ By default, Qwen ignores unmentioned group messages and does not store them as session turns. To let the next `@mention` include recent group context, set `groupHistoryLimit` to a positive number.
177
+
178
+ ```json
179
+ {
180
+ "channels": {
181
+ "my-dingtalk": {
182
+ "type": "dingtalk",
183
+ "clientId": "$DINGTALK_CLIENT_ID",
184
+ "clientSecret": "$DINGTALK_CLIENT_SECRET",
185
+ "groupPolicy": "open",
186
+ "groupHistoryLimit": 50,
187
+ "groups": {
188
+ "*": { "requireMention": true },
189
+ "sensitive-group-id": {
190
+ "requireMention": true,
191
+ "groupHistoryLimit": 0
192
+ }
193
+ }
194
+ }
195
+ }
196
+ }
197
+ ```
198
+
199
+ - Omitted or `0` disables backfill.
200
+ - Group-level `groupHistoryLimit` overrides the channel-level value.
201
+ - Only messages from authorized senders are persisted.
202
+ - Messages rejected by `groupPolicy` or group allowlist are not persisted.
203
+ - Pending group history is stored as local JSONL under `~/.qwen/channels/<channel-name>-group-history.jsonl` or `$QWEN_HOME/channels/<channel-name>-group-history.jsonl`.
204
+ - Cached messages are injected as untrusted context on the next real trigger and are not written as standalone session turns.
205
+
161
206
  ### How group messages are evaluated
162
207
 
163
208
  ```
@@ -312,6 +357,24 @@ qwen channel stop
312
357
 
313
358
  The bot runs in the foreground. Press `Ctrl+C` to stop, or use `qwen channel stop` from another terminal.
314
359
 
360
+ ### Experimental Daemon-Managed Mode
361
+
362
+ You can also run configured channels under `qwen serve`:
363
+
364
+ ```bash
365
+ # Start one channel under the daemon lifecycle
366
+ qwen serve --channel my-channel
367
+
368
+ # Start all configured channels
369
+ qwen serve --channel all
370
+ ```
371
+
372
+ This mode starts one channel worker process owned by `qwen serve`. The worker connects back to the daemon through the SDK and uses the same channel adapters. It is separate from the daemon process, so a channel adapter crash does not crash the daemon.
373
+
374
+ `qwen serve --channel` is not the same service as `qwen channel start`. Standalone `qwen channel start` still uses the ACP-backed channel service and can run channel configs with different `cwd` values. Daemon-managed channels require every selected channel's `cwd` to resolve to the daemon workspace.
375
+
376
+ When channels are serve-managed, `qwen channel status` shows the owner as `qwen serve`, and `qwen channel stop` tells you to stop the daemon instead of signaling the worker directly. If a ready worker exits unexpectedly, the daemon continues running and reports a channel-worker warning in `/daemon/status`.
377
+
315
378
  ### Multi-Channel Mode
316
379
 
317
380
  When you run `qwen channel start` without a name, all channels defined in `settings.json` start together sharing a single agent process. Each channel maintains its own sessions — a Telegram user and a WeChat user get separate conversations, even though they share the same agent.
@@ -11,7 +11,7 @@ Channel plugins are loaded at startup from active extensions. When `qwen channel
11
11
  3. Registers the channel type so it can be referenced in `settings.json`
12
12
  4. Creates channel instances using the plugin's factory function
13
13
 
14
- Your custom channel gets the full shared pipeline for free: sender gating, group policies, session routing, slash commands, crash recovery, and the ACP bridge to the agent.
14
+ Your custom channel gets the full shared pipeline for free: sender gating, group policies, session routing, slash commands, crash recovery, and an agent bridge. Standalone `qwen channel start` currently supplies `AcpBridge`; plugin adapter code should depend on the adapter-facing `ChannelAgentBridge` contract. Existing TypeScript plugins with an explicit `AcpBridge` bridge parameter should migrate that annotation to `ChannelAgentBridge`; JavaScript plugins are unaffected at runtime.
15
15
 
16
16
  ## Installing a Custom Channel
17
17