{ "release": "review-scope-20260930", "candidateRoot": "C:/kefu", "backendFiles": [ "wechat_rpa/admin_backend.py", "wechat_rpa/admin_api.py" ], "frontendDir": "admin-web/apps/web-antd/dist", "frontendSource": "C:/wechat_rpa/admin-web/apps/web-antd/dist", "frontendReplaced": [ "index.html" ], "frontendAdded": [ "js/about-DU6maKCy.js", "js/bootstrap-Bjt0RquF.js", "js/desktop-config-Bzli0qCo.js", "jse/index-index-B_BkXRMk.js" ], "frontendChanged": [ "js/about.js", "js/desktop-config.js" ], "serverTestModules": [ "test_admin_api.ReviewRuleConfigApiTest", "test_admin_api.ReviewRuleValidationTest", "test_admin_api.ConfigNullPoisoningTest" ], "imageTestModules": [ "test_archive_api", "test_zyt_patient_client" ], "schemaMigration": "none", "restartServices": [ "api" ], "notes": "User request 2026-09-30 ('做成管理后台统一设置审核范围' + '客户端也可以设置是否配置'): new desktop config key AI_REVIEW_EVERY_REPLY (bool, default False = original rule: a conversation resumes automatic replies after a verified manual reply; True = review every reply in 审核后发送). Saved from 客户端配置 and shipped with /api/v2/desktop/config; desktops >=1.5.3 follow it unless the PC picks its own scope. admin_backend.py = production (review-gate payload) + this change (3-way merge); admin_api.py = kefu working tree (production == HEAD). web-antd rebuilt from C:/wechat_rpa/admin-web: only desktop-config and the about build time differ in content. Only the api service is recreated; gateway and knowledge-worker keep running. No database change." }