deerflow2/.planning/phases/07-phase-06-mention-upload/07-VALIDATION.md

3.1 KiB
Raw Blame History

phase slug status nyquist_compliant wave_0_complete created
07 phase-06-mention-upload verified true true 2026-04-17

Phase 07 — Validation Strategy

Per-phase validation contract for feedback sampling during execution.


Test Infrastructure

Property Value
Framework Vitest + Playwrightfrontend
Config file frontend/vitest.config.ts, frontend/playwright.config.ts
Quick run command cd frontend && pnpm -s test -- --run src/core/threads
Full suite command `cd frontend && pnpm -s lint && pnpm -s typecheck && pnpm -s test:e2e --grep "input
Estimated runtime ~240 seconds

Sampling Rate

  • After every task commit: Run cd frontend && pnpm -s test -- --run src/core/threads
  • After every plan wave: Run cd frontend && pnpm -s lint && pnpm -s typecheck
  • Before /gsd-verify-work: Full suite must be green
  • Max feedback latency: 300 seconds

Per-Task Verification Map

Task ID Plan Wave Requirement Threat Ref Secure Behavior Test Type Automated Command File Exists Status
07-01-01 01 1 P7-01, P7-02 T-07-01 发送前拼接且消息区不回显拼接文案 unit + e2e node --test frontend/src/core/threads/hooks.test.ts + cd frontend && pnpm -s test:e2e --grep "DF-INPUT-008A" green
07-01-02 01 1 P7-03 T-07-02 附件/Skill 名来源、顺序与去重规则一致 unit node --test frontend/src/core/threads/hooks.test.ts green
07-01-03 01 1 P7-04 T-07-03 所有发送入口行为一致,不出现分叉 e2e `cd frontend && pnpm -s test:e2e --grep "DF-INPUT-003 DF-INPUT-005 DF-INPUT-008A"`

Status: pending · green · red · ⚠️ flaky


Wave 0 Requirements

  • frontend/src/core/threads/hooks.test.ts — 已覆盖提交态增强文本、顺序与去重断言
  • frontend/src/components/workspace/input-box.test.tsx — 由 E2E 发送入口链路覆盖,无独立缺口
  • frontend/tests/e2e/input-and-compose.spec.ts — 已包含“消息区不显示拼接文案”回归DF-INPUT-008A

Manual-Only Verifications

Behavior Requirement Why Manual Test Instructions
多语言文案下拼接语句可读性 P7-01 文案自然性主观 在中文/英文 UI 下分别发送含附件+Skill消息人工检查生成文本

Validation Sign-Off

  • All tasks have <automated> verify or Wave 0 dependencies
  • Sampling continuity: no 3 consecutive tasks without automated verify
  • Wave 0 covers all MISSING references
  • No watch-mode flags
  • Feedback latency < 300s
  • nyquist_compliant: true set in frontmatter

Approval: verified 2026-04-17


Validation Audit 2026-04-17

Metric Count
Gaps found 0
Resolved 0
Escalated 0