lobster-full-engine-validation-20260628-104610.json 82 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847
  1. {
  2. "companyId": 338,
  3. "includeTravelChain": false,
  4. "includeTravelStress": false,
  5. "startedAt": "2026-06-28 10:46:11",
  6. "sections": {
  7. "engineHealth": {
  8. "passed": true,
  9. "health": {
  10. "companyId": 338,
  11. "healthy": true,
  12. "missingCore": [
  13. ],
  14. "missingOptional": [
  15. ],
  16. "passThreshold": 120
  17. },
  18. "message": "core beans wired"
  19. },
  20. "integrationTests": {
  21. "passed": false,
  22. "summary": "2/4 tests passed in 72873ms",
  23. "details": [
  24. "[FAIL] NodeTypeService: START node not found",
  25. "[FAIL] DynamicNodeExecutor: Judgment node execution failed",
  26. "[PASS] MultiModelWorkflowGenerator: All tests passed",
  27. "[PASS] ModelConfigService: All tests passed"
  28. ],
  29. "executionTimeMs": 72873,
  30. "message": "2/4 tests passed in 72873ms"
  31. },
  32. "phase2Replay": {
  33. "passed": true,
  34. "verify": {
  35. "companyId": 338,
  36. "archiveTableExists": true,
  37. "hotIndexExists": true,
  38. "hotReplayRows": 82,
  39. "eventLogRows": 82,
  40. "archiveRows": 0,
  41. "status": "ok"
  42. },
  43. "message": "phase2 replay verify ok"
  44. },
  45. "pageLinkProbes": {
  46. "totalPages": 29,
  47. "requiredTotal": 15,
  48. "requiredPassed": 15,
  49. "optionalSkipped": 0,
  50. "failures": [
  51. ],
  52. "passed": true,
  53. "durationMs": 4946,
  54. "pages": [
  55. {
  56. "path": "/lobster/company-workflowLobster",
  57. "title": "Workflow templates",
  58. "probe": "templateList",
  59. "required": true,
  60. "passed": true,
  61. "status": "OK",
  62. "message": "templates=50"
  63. },
  64. {
  65. "path": "/lobster/workflow-execution",
  66. "title": "Workflow execution tasks",
  67. "probe": "instanceList",
  68. "required": true,
  69. "passed": true,
  70. "status": "OK",
  71. "message": "instances=644"
  72. },
  73. {
  74. "path": "/lobster/instance",
  75. "title": "Instance monitor",
  76. "probe": "instanceMonitor",
  77. "required": true,
  78. "passed": true,
  79. "status": "OK",
  80. "message": "monitor list ok, count=644"
  81. },
  82. {
  83. "path": "/lobster/prompt",
  84. "title": "Prompt management",
  85. "probe": "promptList",
  86. "required": true,
  87. "passed": true,
  88. "status": "OK",
  89. "message": "prompt rows=66"
  90. },
  91. {
  92. "path": "/lobster/sensitive-words",
  93. "title": "Sensitive words",
  94. "probe": "sensitiveWords",
  95. "required": true,
  96. "passed": true,
  97. "status": "OK",
  98. "message": "words=151"
  99. },
  100. {
  101. "path": "/lobster/forbidden-words",
  102. "title": "Forbidden words",
  103. "probe": "forbiddenWords",
  104. "required": false,
  105. "passed": true,
  106. "status": "OK",
  107. "message": "forbidden=32"
  108. },
  109. {
  110. "path": "/lobster/typo-words",
  111. "title": "Typo words",
  112. "probe": "typoWords",
  113. "required": false,
  114. "passed": true,
  115. "status": "OK",
  116. "message": "typo=3"
  117. },
  118. {
  119. "path": "/lobster/conversation-summary",
  120. "title": "Conversation summary",
  121. "probe": "conversationSummary",
  122. "required": true,
  123. "passed": true,
  124. "status": "OK",
  125. "message": "summaries=20"
  126. },
  127. {
  128. "path": "/lobster/profile-config",
  129. "title": "Profile config",
  130. "probe": "profileConfig",
  131. "required": false,
  132. "passed": true,
  133. "status": "OK",
  134. "message": "profile fields=22"
  135. },
  136. {
  137. "path": "/lobster/channel-config",
  138. "title": "Channel config",
  139. "probe": "channelTypes",
  140. "required": true,
  141. "passed": true,
  142. "status": "OK",
  143. "message": "channels=14"
  144. },
  145. {
  146. "path": "/lobster/channel-plugin",
  147. "title": "Channel plugins",
  148. "probe": "channelPlugins",
  149. "required": true,
  150. "passed": true,
  151. "status": "OK",
  152. "message": "plugins=3"
  153. },
  154. {
  155. "path": "/lobster/api-registry",
  156. "title": "API registry",
  157. "probe": "apiRegistry",
  158. "required": true,
  159. "passed": true,
  160. "status": "OK",
  161. "message": "apis=0"
  162. },
  163. {
  164. "path": "/lobster/sales-corpus",
  165. "title": "Sales corpus",
  166. "probe": "salesCorpus",
  167. "required": false,
  168. "passed": false,
  169. "status": "WARN",
  170. "message": "\r\n### Error querying database. Cause: java.sql.SQLSyntaxErrorException: Unknown column \u0027corpus_kind\u0027 in \u0027where clause\u0027\r\n### The error may exist in URL [jar:file:/D:/ylrz_saas_new/java/fs-saas-company/target/fs-saas-company.jar!/BOOT-INF/lib/fs-service-1.1.0.jar!/mapper/company/LobsterSalesCorpusMapper.xml]\r\n### The error may involve defaultParameterMap\r\n### The error occurred while setting parameters\r\n### SQL: select count(*) from lobster_sales_corpus where company_id = ? and (corpus_kind is null or corpus_kind = \u0027\u0027 or corpus_kind = \u0027sales_champion\u0027)\r\n### Cause: java.sql.SQLSyntaxErrorException: Unknown column \u0027corpus_kind\u0027 in \u0027where clause\u0027\n; bad SQL grammar []; nested exception is java.sql.SQLSyntaxErrorException: Unknown column \u0027corpus_kind\u0027 in \u0027where clause\u0027"
  171. },
  172. {
  173. "path": "/lobster/chat-aggregate",
  174. "title": "Chat aggregate",
  175. "probe": "chatAggregate",
  176. "required": false,
  177. "passed": true,
  178. "status": "OK",
  179. "message": "sessions=5"
  180. },
  181. {
  182. "path": "/lobster/industry-bootstrap",
  183. "title": "Industry bootstrap",
  184. "probe": "industryPacks",
  185. "required": true,
  186. "passed": true,
  187. "status": "OK",
  188. "message": "packs=3"
  189. },
  190. {
  191. "path": "/lobster/identity-config",
  192. "title": "External role config",
  193. "probe": "identityConfig",
  194. "required": false,
  195. "passed": true,
  196. "status": "OK",
  197. "message": "identity config loaded"
  198. },
  199. {
  200. "path": "/lobster/semantic-keywords",
  201. "title": "Semantic keywords",
  202. "probe": "semanticKeywords",
  203. "required": true,
  204. "passed": true,
  205. "status": "OK",
  206. "message": "keywords=0"
  207. },
  208. {
  209. "path": "/lobster/optimization",
  210. "title": "AI optimization audit",
  211. "probe": "optimization",
  212. "required": true,
  213. "passed": true,
  214. "status": "OK",
  215. "message": "optimization stats=5"
  216. },
  217. {
  218. "path": "/lobster/dynamic-impl",
  219. "title": "Dynamic node approval",
  220. "probe": "dynamicImpl",
  221. "required": true,
  222. "passed": true,
  223. "status": "OK",
  224. "message": "pending impls=0"
  225. },
  226. {
  227. "path": "/lobster/dashboard",
  228. "title": "Engine dashboard",
  229. "probe": "dashboard",
  230. "required": true,
  231. "passed": true,
  232. "status": "OK",
  233. "message": "dashboard keys=8"
  234. },
  235. {
  236. "path": "/lobster/token-stats",
  237. "title": "Token stats",
  238. "probe": "tokenStats",
  239. "required": false,
  240. "passed": true,
  241. "status": "OK",
  242. "message": "token summary loaded"
  243. },
  244. {
  245. "path": "/lobster/evolution",
  246. "title": "Evolution engine",
  247. "probe": "evolution",
  248. "required": true,
  249. "passed": true,
  250. "status": "OK",
  251. "message": "evolution metrics=6"
  252. },
  253. {
  254. "path": "/lobster/chat-test",
  255. "title": "Chat test",
  256. "probe": "chatTest",
  257. "required": false,
  258. "passed": true,
  259. "status": "OK",
  260. "message": "e2e test service ready"
  261. },
  262. {
  263. "path": "/lobster/ab-test",
  264. "title": "AB / GEPA optimization",
  265. "probe": "abTest",
  266. "required": false,
  267. "passed": true,
  268. "status": "OK",
  269. "message": "active ab tests=0"
  270. },
  271. {
  272. "path": "/lobster/profile",
  273. "title": "Customer profile hub",
  274. "probe": "userProfile",
  275. "required": false,
  276. "passed": true,
  277. "status": "OK",
  278. "message": "profiles=63, domain=true"
  279. },
  280. {
  281. "path": "/lobster/profile-domain",
  282. "title": "Unified profile domain",
  283. "probe": "profileDomain",
  284. "required": false,
  285. "passed": true,
  286. "status": "OK",
  287. "message": "profile domain ready, mode=async"
  288. },
  289. {
  290. "path": "/lobster/qdrant",
  291. "title": "Vector knowledge base",
  292. "probe": "qdrant",
  293. "required": false,
  294. "passed": true,
  295. "status": "OK",
  296. "message": "vectors=0"
  297. },
  298. {
  299. "path": "/lobster/global-stats",
  300. "title": "Global evolution stats",
  301. "probe": "globalStats",
  302. "required": false,
  303. "passed": true,
  304. "status": "OK",
  305. "message": "global metrics=6"
  306. },
  307. {
  308. "path": "/lobster/event-audit",
  309. "title": "Event node audit",
  310. "probe": "eventAudit",
  311. "required": false,
  312. "passed": true,
  313. "status": "OK",
  314. "message": "audits=0"
  315. }
  316. ]
  317. },
  318. "travelFullChain": {
  319. "passed": true,
  320. "skipped": true,
  321. "message": "includeTravelChain=false"
  322. },
  323. "profileDomain": {
  324. "basic": {
  325. "enabled": true,
  326. "projectionMode": "async",
  327. "searchMode": "db",
  328. "esEnabled": false,
  329. "mqConsumerEnabled": true,
  330. "missingBeans": [
  331. ],
  332. "beansOk": true,
  333. "simulationMode": "basic",
  334. "runtime": {
  335. "unifiedSearchTotal": 44,
  336. "unifiedSearchMode": "db",
  337. "searchTotal": 44,
  338. "searchMode": "db",
  339. "identityCount": 44,
  340. "orderStatsSample": 0,
  341. "qwContactSample": 0
  342. },
  343. "passed": true,
  344. "message": "profile domain ready"
  345. },
  346. "fullSimulation": {
  347. "enabled": true,
  348. "projectionMode": "async",
  349. "searchMode": "db",
  350. "esEnabled": false,
  351. "mqConsumerEnabled": true,
  352. "missingBeans": [
  353. ],
  354. "beansOk": true,
  355. "simulationMode": "full",
  356. "runtime": {
  357. "unifiedSearchTotal": 44,
  358. "unifiedSearchMode": "db",
  359. "searchTotal": 44,
  360. "searchMode": "db",
  361. "identityCount": 44,
  362. "orderStatsSample": 0,
  363. "qwContactSample": 0
  364. },
  365. "simulation": {
  366. "patchAdapter": true,
  367. "writePath": {
  368. "published": true,
  369. "readBack": true
  370. },
  371. "backfillLegacy": {
  372. "type": "legacy",
  373. "page": 1,
  374. "size": 5,
  375. "processed": 5,
  376. "published": 5
  377. },
  378. "backfillOrders": {
  379. "type": "order",
  380. "page": 1,
  381. "size": 5,
  382. "processed": 0,
  383. "published": 0
  384. },
  385. "backfillQwContacts": {
  386. "type": "qw_contact",
  387. "page": 1,
  388. "size": 5,
  389. "processed": 0,
  390. "published": 0
  391. }
  392. },
  393. "passed": true,
  394. "message": "profile domain ready"
  395. },
  396. "sourceHooks": {
  397. "companyId": 338,
  398. "simulationMode": "source-hooks",
  399. "sources": [
  400. {
  401. "source": "wx_call",
  402. "eventCount": 1,
  403. "eventOk": true,
  404. "shardId": 18,
  405. "profileKey": "__val_src_wx_call__",
  406. "channelType": "WX",
  407. "sourceVersion": "val:wx_call:1782614850539",
  408. "readBack": true,
  409. "passed": true
  410. },
  411. {
  412. "source": "qw_voice",
  413. "eventCount": 1,
  414. "eventOk": true,
  415. "shardId": 3,
  416. "profileKey": "__val_src_qw_voice__",
  417. "channelType": "QW",
  418. "sourceVersion": "val:qw_voice:1782614850539",
  419. "readBack": true,
  420. "passed": true
  421. },
  422. {
  423. "source": "lobster_chat",
  424. "eventCount": 1,
  425. "eventOk": true,
  426. "shardId": 3,
  427. "profileKey": "__val_src_lobster_chat__",
  428. "channelType": "QW",
  429. "sourceVersion": "val:lobster_chat:1782614850539",
  430. "readBack": true,
  431. "passed": true
  432. },
  433. {
  434. "source": "crm_analyze",
  435. "eventCount": 1,
  436. "eventOk": true,
  437. "shardId": 18,
  438. "profileKey": "__val_src_crm_analyze__",
  439. "channelType": "QW",
  440. "sourceVersion": "val:crm_analyze:1782614850539",
  441. "readBack": true,
  442. "passed": true
  443. },
  444. {
  445. "source": "qw_ai_analyze",
  446. "eventCount": 1,
  447. "eventOk": true,
  448. "shardId": 11,
  449. "profileKey": "__val_src_qw_ai_analyze__",
  450. "channelType": "QW",
  451. "sourceVersion": "val:qw_ai_analyze:1782614850539",
  452. "readBack": true,
  453. "passed": true
  454. },
  455. {
  456. "source": "qw_contact",
  457. "eventCount": 1,
  458. "eventOk": true,
  459. "shardId": 14,
  460. "profileKey": "__val_src_qw_contact__",
  461. "channelType": "QW",
  462. "sourceVersion": "val:qw_contact:1782614850539",
  463. "readBack": true,
  464. "passed": true
  465. },
  466. {
  467. "source": "qw_tag",
  468. "eventCount": 1,
  469. "eventOk": true,
  470. "shardId": 27,
  471. "profileKey": "__val_src_qw_tag__",
  472. "channelType": "QW",
  473. "sourceVersion": "val:qw_tag:1782614850539",
  474. "readBack": true,
  475. "passed": true
  476. },
  477. {
  478. "source": "order",
  479. "eventCount": 1,
  480. "eventOk": true,
  481. "shardId": 29,
  482. "profileKey": "__val_src_order__",
  483. "channelType": "QW",
  484. "sourceVersion": "val:order:1782614850539",
  485. "readBack": true,
  486. "passed": true
  487. },
  488. {
  489. "source": "course_watch",
  490. "eventCount": 1,
  491. "eventOk": true,
  492. "shardId": 20,
  493. "profileKey": "__val_src_course_watch__",
  494. "channelType": "QW",
  495. "sourceVersion": "val:course_watch:1782614850539",
  496. "readBack": true,
  497. "passed": true
  498. },
  499. {
  500. "source": "live_watch",
  501. "variant": "meta",
  502. "eventCount": 1,
  503. "eventOk": true,
  504. "shardId": 15,
  505. "profileKey": "__val_src_live_watch__",
  506. "channelType": "QW",
  507. "sourceVersion": "val:live:1782614850539:meta",
  508. "readBack": true,
  509. "passed": true
  510. },
  511. {
  512. "source": "live_watch",
  513. "variant": "counter",
  514. "eventCount": 1,
  515. "eventOk": true,
  516. "shardId": 15,
  517. "profileKey": "__val_src_live_watch__",
  518. "channelType": "QW",
  519. "sourceVersion": "val:live:1782614850539:cnt:1",
  520. "readBack": true,
  521. "passed": true
  522. },
  523. {
  524. "source": "sop",
  525. "eventCount": 1,
  526. "eventOk": true,
  527. "shardId": 20,
  528. "profileKey": "__val_src_sop__",
  529. "channelType": "QW",
  530. "sourceVersion": "val:sop:1782614850539",
  531. "readBack": true,
  532. "passed": true
  533. }
  534. ],
  535. "passedCount": 12,
  536. "totalCount": 12,
  537. "passed": true,
  538. "message": "all source hooks simulated and verified"
  539. },
  540. "passed": true,
  541. "message": "profile domain all validations passed"
  542. },
  543. "semanticFull": {
  544. "companyId": 338,
  545. "sections": {
  546. "semanticKeywordContext": {
  547. "passed": false,
  548. "caseCount": 8,
  549. "message": "soft_refusal_first expected L2 without exclude; soft_refusal_second expected L4 exclude on 2nd hit; negation_soft_refusal should not trigger soft refusal"
  550. },
  551. "catalogValidation": {
  552. "companyId": 338,
  553. "categories": {
  554. "custom_keywords": {
  555. "keywordCount": 129,
  556. "totalCases": 258,
  557. "passedCases": 258,
  558. "passed": true,
  559. "failures": [
  560. ],
  561. "failureSampleCount": 0,
  562. "message": "all 258 checks passed for 129 words"
  563. },
  564. "positive_custom": {
  565. "keywordCount": 107,
  566. "totalCases": 214,
  567. "passedCases": 214,
  568. "passed": true,
  569. "failures": [
  570. ],
  571. "failureSampleCount": 0,
  572. "message": "all 214 checks passed for 107 words"
  573. },
  574. "negative_custom": {
  575. "keywordCount": 108,
  576. "totalCases": 216,
  577. "passedCases": 216,
  578. "passed": true,
  579. "failures": [
  580. ],
  581. "failureSampleCount": 0,
  582. "message": "all 216 checks passed for 108 words"
  583. },
  584. "soft_refusal": {
  585. "keywordCount": 56,
  586. "totalCases": 168,
  587. "passedCases": 112,
  588. "passed": false,
  589. "failures": [
  590. "soft_refusal:下个月 L2 first",
  591. "soft_refusal:以后再说 L2 first",
  592. "soft_refusal:暂时不需要 L2 first",
  593. "soft_refusal:过段时间 L2 first",
  594. "soft_refusal:下次吧 L2 first",
  595. "soft_refusal:改天 L2 first",
  596. "soft_refusal:回头再说 L2 first",
  597. "soft_refusal:再看看 L2 first",
  598. "soft_refusal:再考虑 L2 first",
  599. "soft_refusal:考虑考虑 L2 first",
  600. "soft_refusal:商量一下 L2 first",
  601. "soft_refusal:问问家人 L2 first",
  602. "soft_refusal:问问朋友 L2 first",
  603. "soft_refusal:问问领导 L2 first",
  604. "soft_refusal:汇报一下 L2 first",
  605. "soft_refusal:审批一下 L2 first",
  606. "soft_refusal:走流程 L2 first",
  607. "soft_refusal:需要审批 L2 first",
  608. "soft_refusal:需要汇报 L2 first",
  609. "soft_refusal:需要请示 L2 first",
  610. "soft_refusal:需要讨论 L2 first",
  611. "soft_refusal:开会研究 L2 first",
  612. "soft_refusal:现在不方便 L2 first",
  613. "soft_refusal:在忙 L2 first",
  614. "soft_refusal:在开会 L2 first",
  615. "soft_refusal:在开车 L2 first",
  616. "soft_refusal:稍后再说 L2 first",
  617. "soft_refusal:晚点再说 L2 first",
  618. "soft_refusal:有空再说 L2 first",
  619. "soft_refusal:等一下 L2 first",
  620. "soft_refusal:等会儿 L2 first",
  621. "soft_refusal:待定 L2 first",
  622. "soft_refusal:再议 L2 first",
  623. "soft_refusal:看看再说 L2 first",
  624. "soft_refusal:想想再说 L2 first",
  625. "soft_refusal:再想想 L2 first",
  626. "soft_refusal:暂时算了 L2 first",
  627. "soft_refusal:先这样 L2 first",
  628. "soft_refusal:以后再聊 L2 first",
  629. "soft_refusal:以后再联系 L2 first"
  630. ],
  631. "failureSampleCount": 56,
  632. "message": "56 failed of 168"
  633. },
  634. "handoff_trigger": {
  635. "keywordCount": 78,
  636. "totalCases": 390,
  637. "passedCases": 390,
  638. "passed": true,
  639. "failures": [
  640. ],
  641. "failureSampleCount": 0,
  642. "message": "all 390 checks passed for 78 words"
  643. },
  644. "sarcasm": {
  645. "keywordCount": 61,
  646. "totalCases": 122,
  647. "passedCases": 122,
  648. "passed": true,
  649. "failures": [
  650. ],
  651. "failureSampleCount": 0,
  652. "message": "all 122 checks passed for 61 words"
  653. },
  654. "_combos": {
  655. "totalCases": 6,
  656. "passedCases": 5,
  657. "passed": false,
  658. "failures": [
  659. "combo:soft_refusal 下个月,需要汇报,暂时算了"
  660. ],
  661. "message": "combo:soft_refusal 下个月,需要汇报,暂时算了"
  662. }
  663. },
  664. "categoryCount": 6,
  665. "totalCases": 1374,
  666. "passedCases": 1317,
  667. "failedCases": 57,
  668. "failures": [
  669. "soft_refusal:下个月 L2 first",
  670. "soft_refusal:以后再说 L2 first",
  671. "soft_refusal:暂时不需要 L2 first",
  672. "soft_refusal:过段时间 L2 first",
  673. "soft_refusal:下次吧 L2 first",
  674. "soft_refusal:改天 L2 first",
  675. "soft_refusal:回头再说 L2 first",
  676. "soft_refusal:再看看 L2 first",
  677. "soft_refusal:再考虑 L2 first",
  678. "soft_refusal:考虑考虑 L2 first",
  679. "soft_refusal:商量一下 L2 first",
  680. "soft_refusal:问问家人 L2 first",
  681. "soft_refusal:问问朋友 L2 first",
  682. "soft_refusal:问问领导 L2 first",
  683. "soft_refusal:汇报一下 L2 first",
  684. "soft_refusal:审批一下 L2 first",
  685. "soft_refusal:走流程 L2 first",
  686. "soft_refusal:需要审批 L2 first",
  687. "soft_refusal:需要汇报 L2 first",
  688. "soft_refusal:需要请示 L2 first",
  689. "soft_refusal:需要讨论 L2 first",
  690. "soft_refusal:开会研究 L2 first",
  691. "soft_refusal:现在不方便 L2 first",
  692. "soft_refusal:在忙 L2 first",
  693. "soft_refusal:在开会 L2 first",
  694. "soft_refusal:在开车 L2 first",
  695. "soft_refusal:稍后再说 L2 first",
  696. "soft_refusal:晚点再说 L2 first",
  697. "soft_refusal:有空再说 L2 first",
  698. "soft_refusal:等一下 L2 first",
  699. "soft_refusal:等会儿 L2 first",
  700. "soft_refusal:待定 L2 first",
  701. "soft_refusal:再议 L2 first",
  702. "soft_refusal:看看再说 L2 first",
  703. "soft_refusal:想想再说 L2 first",
  704. "soft_refusal:再想想 L2 first",
  705. "soft_refusal:暂时算了 L2 first",
  706. "soft_refusal:先这样 L2 first",
  707. "soft_refusal:以后再聊 L2 first",
  708. "soft_refusal:以后再联系 L2 first"
  709. ],
  710. "failureTruncated": true,
  711. "passed": false,
  712. "durationMs": 125130,
  713. "message": "41 failure(s), showing up to 40"
  714. }
  715. },
  716. "contextCases": [
  717. {
  718. "id": "soft_refusal_first",
  719. "input": "下个月再说 (empty ctx)",
  720. "expected": "L2 pause, not exclude",
  721. "actual": "null exclude=false reason=null",
  722. "passed": false
  723. },
  724. {
  725. "id": "soft_refusal_second",
  726. "input": "下个月再说 (_softRefusalCount=1)",
  727. "expected": "L4 exclude (repeated)",
  728. "actual": "null exclude=false reason=null",
  729. "passed": false
  730. },
  731. {
  732. "id": "handoff_first_neutral",
  733. "input": "了解一下人工 -\u003e keyword 人工 (neutral)",
  734. "expected": "no transfer, count=1",
  735. "actual": "takeover=false count=1 reason=Handoff trigger noted (session 1, awaiting confirmation): 人工",
  736. "passed": true
  737. },
  738. {
  739. "id": "handoff_second_confirm",
  740. "input": "转人工 (_handoffTriggerCount=1)",
  741. "expected": "transfer on 2nd hit",
  742. "actual": "takeover=true reason=Handoff trigger (session escalated x2): 人工",
  743. "passed": true
  744. },
  745. {
  746. "id": "negation_handoff",
  747. "input": "不转人工",
  748. "expected": "no handoff",
  749. "actual": "takeover=false",
  750. "passed": true
  751. },
  752. {
  753. "id": "negation_soft_refusal",
  754. "input": "别下个月",
  755. "expected": "no soft refusal",
  756. "actual": "L4_EXCLUDE reason=negative_intent",
  757. "passed": false
  758. },
  759. {
  760. "id": "positive_sentiment_soft",
  761. "input": "很满意,价格多少 (positive ctx)",
  762. "expected": "no soft refusal",
  763. "actual": "null exclude=false",
  764. "passed": true
  765. },
  766. {
  767. "id": "positive_sentiment_handoff",
  768. "input": "很满意,价格多少 (positive ctx, no handoff keyword)",
  769. "expected": "no handoff misfire",
  770. "actual": "takeover=false reason=客户表达了满意的态度,诉求是询问价格,属于常规的咨询类问题,无需专业/专项帮助",
  771. "passed": true
  772. }
  773. ],
  774. "failures": [
  775. "soft_refusal_first expected L2 without exclude",
  776. "soft_refusal_second expected L4 exclude on 2nd hit",
  777. "negation_soft_refusal should not trigger soft refusal",
  778. "catalog: 41 failure(s), showing up to 40"
  779. ],
  780. "passed": false,
  781. "durationMs": 137472,
  782. "totalCases": 1382,
  783. "passedCases": 1322,
  784. "message": "soft_refusal_first expected L2 without exclude; soft_refusal_second expected L4 exclude on 2nd hit; negation_soft_refusal should not trigger soft refusal; catalog: 41 failure(s), showing up to 40",
  785. "catalogFailureSamples": [
  786. "soft_refusal:下个月 L2 first",
  787. "soft_refusal:以后再说 L2 first",
  788. "soft_refusal:暂时不需要 L2 first",
  789. "soft_refusal:过段时间 L2 first",
  790. "soft_refusal:下次吧 L2 first",
  791. "soft_refusal:改天 L2 first",
  792. "soft_refusal:回头再说 L2 first",
  793. "soft_refusal:再看看 L2 first",
  794. "soft_refusal:再考虑 L2 first",
  795. "soft_refusal:考虑考虑 L2 first",
  796. "soft_refusal:商量一下 L2 first",
  797. "soft_refusal:问问家人 L2 first",
  798. "soft_refusal:问问朋友 L2 first",
  799. "soft_refusal:问问领导 L2 first",
  800. "soft_refusal:汇报一下 L2 first",
  801. "soft_refusal:审批一下 L2 first",
  802. "soft_refusal:走流程 L2 first",
  803. "soft_refusal:需要审批 L2 first",
  804. "soft_refusal:需要汇报 L2 first",
  805. "soft_refusal:需要请示 L2 first",
  806. "soft_refusal:需要讨论 L2 first",
  807. "soft_refusal:开会研究 L2 first",
  808. "soft_refusal:现在不方便 L2 first",
  809. "soft_refusal:在忙 L2 first",
  810. "soft_refusal:在开会 L2 first",
  811. "soft_refusal:在开车 L2 first",
  812. "soft_refusal:稍后再说 L2 first",
  813. "soft_refusal:晚点再说 L2 first",
  814. "soft_refusal:有空再说 L2 first",
  815. "soft_refusal:等一下 L2 first",
  816. "soft_refusal:等会儿 L2 first",
  817. "soft_refusal:待定 L2 first",
  818. "soft_refusal:再议 L2 first",
  819. "soft_refusal:看看再说 L2 first",
  820. "soft_refusal:想想再说 L2 first",
  821. "soft_refusal:再想想 L2 first",
  822. "soft_refusal:暂时算了 L2 first",
  823. "soft_refusal:先这样 L2 first",
  824. "soft_refusal:以后再聊 L2 first",
  825. "soft_refusal:以后再联系 L2 first"
  826. ]
  827. }
  828. },
  829. "integrationWarning": "2/4 tests passed in 72873ms",
  830. "failures": [
  831. "semanticFull: soft_refusal_first expected L2 without exclude"
  832. ],
  833. "passed": false,
  834. "durationMs": 219308,
  835. "pageCount": 29,
  836. "requiredPagesPassed": "15/15"
  837. }