feat(perf): add no-hardware startup signoff pipeline

Author: Jacobinwwey

docs/diataxis/en/explanation/startup-node-update-acceleration-plan.md

@@ -213,3 +213,11 @@ npm run perf:startup:matrix -- --root tmp/startup-logs --single-platform-label w
 - Phase 4 (verification and rollout gate): automation completed  
   - `compare / matrix / cohorts` gate scripts are available and validated.
   - Release-grade cross-platform signoff still requires real macOS/Android/iOS datasets.
+
+## 15) Signoff Strategy Without Multi-Device Hardware (March 31, 2026)
+
+- One-click command: `npm run perf:startup:signoff:nohw`
+- Output: `tmp/startup-logs/report-startup-signoff.md`
+- Layered decision model:
+  - Engineering signoff: based on Windows real logs + simulated multi-platform three-cohort gates.
+  - Release signoff: marked as `BLOCKED` when real multi-device cohorts are missing (never falsely `PASS`).

docs/diataxis/en/reference/interfaces-and-runtime.md

@@ -82,6 +82,10 @@ This reference tracks canonical API/runtime contracts.
   - `npm run perf:startup:matrix:simulate -- --seed-root tmp/startup-logs --out-root tmp/startup-logs-simulated`
   - `npm run perf:startup:matrix -- --root tmp/startup-logs-simulated --out tmp/startup-logs-simulated/report-platform-matrix.md`
   - Note: `tmp/startup-logs-simulated` is synthetic data and must not be used for release-go performance decisions.
+- One-click no-hardware signoff (engineering gate + release gate blocking status):
+  - `npm run perf:startup:signoff:nohw`
+  - This command automatically runs: Windows real-log gates + simulated multi-platform three-cohort gates, then writes a layered signoff report.
+  - If real multi-device cohort data is missing, release signoff is marked as `BLOCKED` (not falsely `PASS`).
 - Three-cohort automatic regression and rollout gate (Phase 4):
   - `npm run perf:startup:cohorts:verify -- --root <cohorts-root> --cohorts small,medium,large --out <report-path> --strict`
   - Directory contract: `<cohorts-root>/<cohort>/<platform>/baseline|pilot`

docs/diataxis/zh/explanation/startup-node-update-acceleration-plan.md

@@ -213,3 +213,11 @@ npm run perf:startup:matrix -- --root tmp/startup-logs --single-platform-label w
 - Phase 4（验证灰度）：已完成自动化链路  
   - 已有 `compare / matrix / cohorts` 三类门禁脚本。
   - 当前仍需补齐真实多端样本（macOS/Android/iOS）以完成发布级跨端验收。
+
+## 15) 无多端硬件时的签收策略（2026-03-31）
+
+- 一键命令：`npm run perf:startup:signoff:nohw`
+- 产物：`tmp/startup-logs/report-startup-signoff.md`
+- 分层结论：
+  - 工程签收：基于 Windows 真实日志 + 模拟多端三规模门禁。
+  - 发布签收：若缺少真实多端 cohort，状态为 `BLOCKED`（不会误判为 `PASS`）。

docs/diataxis/zh/reference/interfaces-and-runtime.md

@@ -82,6 +82,10 @@
   - `npm run perf:startup:matrix:simulate -- --seed-root tmp/startup-logs --out-root tmp/startup-logs-simulated`
   - `npm run perf:startup:matrix -- --root tmp/startup-logs-simulated --out tmp/startup-logs-simulated/report-platform-matrix.md`
   - 注意：`tmp/startup-logs-simulated` 为模拟数据，禁止用于 release-go 性能结论，仅用于脚本/门禁流程演练。
+- 无硬件一键签收（工程签收 + 发布签收阻塞提示）：
+  - `npm run perf:startup:signoff:nohw`
+  - 该命令会自动执行：Windows 真实日志门禁 + 模拟多端三规模门禁，并输出分层签收报告。
+  - 若缺少真实多端 cohort 数据，发布签收会标记为 `BLOCKED`（而非误判为 `PASS`）。
 - 三规模自动回归与灰度门禁（Phase 4）：
   - `npm run perf:startup:cohorts:verify -- --root <cohorts-root> --cohorts small,medium,large --out <report-path> --strict`
   - 目录约定：`<cohorts-root>/<cohort>/<platform>/baseline|pilot`

package.json

@@ -62,6 +62,7 @@
     "perf:startup:matrix:watch": "node scripts/watch-startup-perf-matrix.js",
     "perf:startup:matrix:simulate": "node scripts/simulate-startup-perf-platform-logs.js",
     "perf:startup:cohorts:verify": "node scripts/verify-startup-perf-cohorts.js",
+    "perf:startup:signoff:nohw": "node scripts/verify-startup-release-readiness.js --strict-engineering",
     "verify:pathbridge:strict": "node scripts/verify-pathbridge-strict-schema.js",
     "generate:sbom": "node scripts/generate-sbom.js",
     "generate:sbom:attestation": "node scripts/generate-sbom-attestation.js",