Artifact kind: e2e_verification.
| 測試 | 方法 | Pass |
|---|---|---|
| 訪前判讀完整 | 檢查 workbench | 有客戶產業、痛點、案例邊界、決策角色 |
| 45 分鐘腳本可執行 | 模擬拜訪 | 每段有目標、話術、要記錄欄位 |
Given an accepted reviewer inbox and 50-case seed queue, when the worker starts a D7 calibration run, then the run must be created with exactly 50 unique calibration_run_item rows and source freshness metadata.
Pass evidence:
calibration_run.status=readytotal_items=50Given D7 calibration and D14 correction artifacts, when D30 weekly scorecard is generated, then it must include run id, correction batch id, evidence snapshot, and decision record reference.
Pass:
calibration_run_id exists.correction_batch_id exists.Given a scorecard payload, when PLS loads /ai-prediction/scorecards/:id/card, then the card must render gate header, metric strip, repair backlog, evidence drawer, and next worker action.
Pass:
ship_weekly_scorecard, repair_first, blocked.Given the project already has D7-D30, dashboard spec, and repo handoff, when PLS creates the next job, then it must not be another generic project_runner production pack.
Pass:
repo_change or github_pr.ai_native_project_id has 2+ completed durable artifacts and no target_repo_url,dispatch_repair_record and block generic project_runner completion as the next lane.Expected evidence: next job payload contains missing_capability=target_repo_url and next_worker_kind=repo_change/github_pr.
Expected evidence: import result includes row_count, accepted rows, rejected rows, and schema error list.