dsh-doublecheck 0.9.26 → 0.9.27

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -22,6 +22,11 @@
22
22
 
23
23
  [English](README.md) · [简体中文](README-zh.md) · [Español](README-es.md) · [Português](README-pt.md) · [हिन्दी](README-hi.md)
24
24
 
25
+ ```sh
26
+ dsh plugin --profile <name> add dsh-doublecheck
27
+ ```
28
+
29
+
25
30
  </div>
26
31
 
27
32
  ---
@@ -29,6 +34,15 @@
29
34
  **📖 Ecosystem knowledge base** — measured data, not marketing: [plugin development guide · plugin-selection data · maintenance criteria](https://perrylink.github.io/dsh-plugin-guide/).
30
35
 
31
36
  <!-- star-cta -->
37
+ ## Install & uninstall
38
+
39
+ - **git channel** (latest `main`): `dsh plugin --profile web add github:PerryLink/dsh-doublecheck` — the `prepare` script builds with production dependencies only.
40
+ - **npm channel** (published releases): `dsh plugin --profile web add dsh-doublecheck`.
41
+ - **tarball channel**: `pnpm pack` in this repo, then `dsh plugin --profile web add ./dsh-doublecheck-<version>.tgz`.
42
+ - **uninstall**: `dsh plugin --profile web remove dsh-doublecheck` (or remove the rows from the profile patch).
43
+
44
+ For a zero-configuration strict mode (every gate on at `block` intensity, gate coverage required), apply the shipped overlay on top of the bundle patch: `dsh --profile web --patch ./node_modules/dsh-doublecheck/strict.patch.yml`.
45
+
32
46
  ## ⭐ 如果它帮到了你
33
47
 
34
48
  这个插件是 [DSH 插件家族](https://github.com/PerryLink)的一员(40+ 个,全部 Apache-2.0)。如果你在用,**给个 star** —— 它不会解锁任何功能,但会让下一个人在搜索里更容易找到它。
@@ -101,15 +115,6 @@ dsh --profile web --dump-config | grep -E -A3 'id: doublecheck-(grill|guard)'
101
115
 
102
116
  Both rows (`doublecheck-grill` and `doublecheck-guard`) activate automatically with the profile.
103
117
 
104
- ## Install & uninstall
105
-
106
- - **git channel** (latest `main`): `dsh plugin --profile web add github:PerryLink/dsh-doublecheck` — the `prepare` script builds with production dependencies only.
107
- - **npm channel** (published releases): `dsh plugin --profile web add dsh-doublecheck`.
108
- - **tarball channel**: `pnpm pack` in this repo, then `dsh plugin --profile web add ./dsh-doublecheck-<version>.tgz`.
109
- - **uninstall**: `dsh plugin --profile web remove dsh-doublecheck` (or remove the rows from the profile patch).
110
-
111
- For a zero-configuration strict mode (every gate on at `block` intensity, gate coverage required), apply the shipped overlay on top of the bundle patch: `dsh --profile web --patch ./node_modules/dsh-doublecheck/strict.patch.yml`.
112
-
113
118
  ## Configuration
114
119
 
115
120
  All tunables are Schemastery `Config` fields (changeable from cordis.yml). An id-targeted override replaces the whole row — restate every key you need. `cordis.patch.yml` documents each key inline; Schema defaults are the single source of tuning defaults.
@@ -344,55 +349,112 @@ Static evidence: `dsh-plugin-doctor` K10–K13 report `pass` for every check on
344
349
 
345
350
  ## PerryLink DSH Plugin Family
346
351
 
347
- This project is one of the **34 actively maintained** DeepSeek Harness plugins from [PerryLink](https://github.com/PerryLink) — the roster is **42**, of which **5** are frozen and **3** retired; every one keeps its row below, with the reason in the Status column. If this one helps you, the others likely will too:
352
+ This project is one of the **87 actively maintained** DeepSeek Harness plugins from [PerryLink](https://github.com/PerryLink). The family roster is **95** plugins in total — **5 frozen**, **3 retired** — and stays collapsed below so every README keeps a short first screen. Expand it to see every row, with the reason in the Status column.
353
+
354
+ <details>
355
+ <summary>Expand all 95 plugins</summary>
348
356
 
349
357
  | Plugin | One-liner | Status |
350
358
  |---|---|---|
351
- | **[dsh-auto-review](https://github.com/PerryLink/dsh-auto-review)** | Second-model auto-review on the approval chain, fail-closed by default | |
352
- | **[dsh-autotier](https://github.com/PerryLink/dsh-autotier)** | Automatic strong/cheap model-tier routing with deterministic risk guards and a `/tier` command | |
353
- | **[dsh-background-agents](https://github.com/PerryLink/dsh-background-agents)** | Durable background child agents with a Web UI sidebar, messaging and interrupt | 🚫 **RETIRED** — see the note above |
359
+ | **[dsh-aqua-input-check](https://github.com/PerryLink/dsh-aqua-input-check)** | 水产养殖投入品记录核对(按投入品用量、休药期与起捕间隔核对记录自洽,仅提示差异,不作出定性结论) | |
360
+ | **[dsh-archive-check](https://github.com/PerryLink/dsh-archive-check)** | 档案归档完整性与保管期限核对(依据《档案法》及配套规定,仅提示差异,不作出定性结论) | |
361
+ | **[dsh-auto-review](https://github.com/PerryLink/dsh-auto-review)** | Second-model auto-review on the approval chain, fail-closed by default | |
362
+ | **[dsh-autotier](https://github.com/PerryLink/dsh-autotier)** | Automatic strong/cheap model-tier routing with deterministic risk guards and a `/tier` command | |
363
+ | **[dsh-background-agents](https://github.com/PerryLink/dsh-background-agents)** | Durable background child agents with a Web UI sidebar, messaging and interrupt | 🚫 RETIRED — see the repo README |
364
+ | **[dsh-bid-ca-precheck](https://github.com/PerryLink/dsh-bid-ca-precheck)** | 投标文件符合性预检(按招标文件逐条核对响应情况,仅提示差异,不作出定性结论) | |
365
+ | **[dsh-bid-qual-check](https://github.com/PerryLink/dsh-bid-qual-check)** | 投标人资格条件核对(按资格条件与证明材料核对逐条响应,仅提示差异,不作出定性结论) | |
354
366
  | **[dsh-budget](https://github.com/PerryLink/dsh-budget)** | Cost governance for DeepSeek Harness: budgets, carbon, and latency in one panel. | 🧊 FROZEN — see the repo README |
355
- | **[dsh-catalog](https://github.com/PerryLink/dsh-catalog)** | DSH Desktop Market standard catalog source for the PerryLink family | |
356
- | **[dsh-cert-mcp](https://github.com/PerryLink/dsh-cert-mcp)** | Read-only MCP server exposing the certification registry: grades, snapshots and five-dimension evidence | |
357
- | **[dsh-checkpoint-rewind](https://github.com/PerryLink/dsh-checkpoint-rewind)** | Claude Code /rewind-equivalent: snapshots, session forks, one-shot restore | |
367
+ | **[dsh-cert-mcp](https://github.com/PerryLink/dsh-cert-mcp)** | Read-only MCP server exposing the certification registry: grades, snapshots and five-dimension evidence | |
368
+ | **[dsh-checkpoint-rewind](https://github.com/PerryLink/dsh-checkpoint-rewind)** | Claude Code /rewind-equivalent: snapshots, session forks, one-shot restore | |
358
369
  | **[dsh-claude-move](https://github.com/PerryLink/dsh-claude-move)** | Migrate Claude Code sessions, memory, skills and CLAUDE.md into DSH | 🧊 FROZEN — see the repo README |
359
- | **[dsh-click](https://github.com/PerryLink/dsh-click)** | Cross-platform native desktop control for DeepSeek Harness — Windows first. | |
360
- | **[dsh-composer-history](https://github.com/PerryLink/dsh-composer-history)** | Terminal-style input history for the web composer: arrows, Ctrl+R search | |
361
- | **[dsh-data-quality](https://github.com/PerryLink/dsh-data-quality)** | Dataset quality checks and citation cross-checks (the optional numeric bridge consumed here) | |
370
+ | **[dsh-click](https://github.com/PerryLink/dsh-click)** | Cross-platform native desktop control for DeepSeek Harness — Windows first. | |
371
+ | **[dsh-composer-history](https://github.com/PerryLink/dsh-composer-history)** | Terminal-style input history for the web composer: arrows, Ctrl+R search | |
372
+ | **[dsh-contract-stance](https://github.com/PerryLink/dsh-contract-stance)** | 合同条款立场台账核对(按立场、风险等级与退让底线核对台账自洽,仅提示差异,不作出定性结论) | |
373
+ | **[dsh-customs-doc-check](https://github.com/PerryLink/dsh-customs-doc-check)** | 报关单证一致性核对(按公开的报关单填制规范核对单证间字段一致性,仅提示差异,不作出定性结论) | |
374
+ | **[dsh-data-quality](https://github.com/PerryLink/dsh-data-quality)** | Dataset quality checks and citation cross-checks (the optional numeric bridge consumed here) | |
362
375
  | **[dsh-defend](https://github.com/PerryLink/dsh-defend)** | Prompt-injection, jailbreak, and secret-leak defense for DeepSeek Harness. | 🧊 FROZEN — see the repo README |
363
- | **[dsh-doublecheck](https://github.com/PerryLink/dsh-doublecheck)** | Engineering-discipline guard: requirements grill, test gates, adversary review | |
376
+ | **[dsh-demurrage-ledger](https://github.com/PerryLink/dsh-demurrage-ledger)** | 滞箱费台账核对(按免箱期与费率核对超期天数与金额的自洽性,仅提示差异,不作出定性结论) | |
377
+ | **[dsh-doublecheck](https://github.com/PerryLink/dsh-doublecheck)** | Engineering-discipline guard: requirements grill, test gates, adversary review | |
364
378
  | **[dsh-draw](https://github.com/PerryLink/dsh-draw)** | Unified static-image generation routing for DeepSeek Harness. | 🧊 FROZEN — see the repo README |
365
- | **[dsh-fast](https://github.com/PerryLink/dsh-fast)** | Read-only performance diagnostics for DeepSeek Harness. | |
366
- | **[dsh-fund-research](https://github.com/PerryLink/dsh-fund-research)** | Deterministic research reports for Chinese public mutual funds | |
367
- | **[dsh-github](https://github.com/PerryLink/dsh-github)** | GitHub PR/issues integration for DSH, every write gated by approval | |
368
- | **[dsh-industry-research](https://github.com/PerryLink/dsh-industry-research)** | Industry research orchestration that seals its deliverables through this plugin's `ctx.researchReport.assemble` | |
369
- | **[dsh-laya](https://github.com/PerryLink/dsh-laya)** | Laya typed decisions (`noul`/`choice`/`score`) as a first-class Cordis service and model-visible tools | |
370
- | **[dsh-library](https://github.com/PerryLink/dsh-library)** | Local document knowledge base for DeepSeek Harness. | |
371
- | **[dsh-local-ai](https://github.com/PerryLink/dsh-local-ai)** | Local-model (Ollama) integration for DeepSeek Harness. | |
372
- | **[dsh-lsp-actions](https://github.com/PerryLink/dsh-lsp-actions)** | LSP diagnostics, formatting, completion, code actions and rename over language servers | |
373
- | **[dsh-mask](https://github.com/PerryLink/dsh-mask)** | PII masking middleware: anonymize at the model boundary, restore at the display layer | |
374
- | **[dsh-mcp-panel](https://github.com/PerryLink/dsh-mcp-panel)** | Read-only MCP runtime panel: /mcp command + Settings tab with status, tools and errors | |
375
- | **[dsh-memento](https://github.com/PerryLink/dsh-memento)** | Approval-gated cross-session memory: ctx.memory seam + SQLite + memory tool | |
376
- | **[dsh-observe](https://github.com/PerryLink/dsh-observe)** | OpenTelemetry and Langfuse observability exporter for DeepSeek Harness. | |
377
- | **[dsh-output-styles](https://github.com/PerryLink/dsh-output-styles)** | Claude Code outputStyles-equivalent runtime style switching | |
378
- | **[dsh-permission-rules](https://github.com/PerryLink/dsh-permission-rules)** | Claude Code-style declarative allow/deny/ask permission rules with audit | |
379
- | **[dsh-plugin-certification](https://github.com/PerryLink/dsh-plugin-certification)** | Community certification registry with repro-checkable grades and badges | |
380
- | **[dsh-plugin-doctor](https://github.com/PerryLink/dsh-plugin-doctor)** | Zero-dependency static + sandbox smoke detector for DSH plugins | |
381
- | **[dsh-plugin-guide](https://github.com/PerryLink/dsh-plugin-guide)** | Plugin-development knowledge base as an on-demand agent skill | |
382
- | **[dsh-plugin-kit](https://github.com/PerryLink/dsh-plugin-kit)** | Shared zero-runtime-dependency toolkit for the PerryLink DSH plugins | |
383
- | **[dsh-plugin-upgrade](https://github.com/PerryLink/dsh-plugin-upgrade)** | One-package, one-corridor-index plugin upgrade skill: routes a repository to the matching closed corridor card | |
384
- | **[dsh-plugin-upgrade-015](https://github.com/PerryLink/dsh-plugin-upgrade-015)** | Merged `0.1.3-alpha.1` → `0.1.5-rc.1` upgrade corridor card plus a zero-dependency seam scanner | 🚫 RETIRED — corridors carried by `dsh-plugin-upgrade` |
379
+ | **[dsh-drill-script-check](https://github.com/PerryLink/dsh-drill-script-check)** | 应急演练脚本核对(按阶段、响应分级、指挥与资源要素核对脚本齐备,仅提示差异,不作出定性结论) | |
380
+ | **[dsh-eia-guide-check](https://github.com/PerryLink/dsh-eia-guide-check)** | 环评导则适用性核对(按要素、导则名称与执行标准核对台账自洽,仅提示差异,不作出定性结论) | |
381
+ | **[dsh-emergency-plan](https://github.com/PerryLink/dsh-emergency-plan)** | 应急预案要素齐备性与条款级页码定位(依据 GB/T 29639-2020 等公开文件,仅提示差异,不作出定性结论) | |
382
+ | **[dsh-essay-rubric-check](https://github.com/PerryLink/dsh-essay-rubric-check)** | 作文量表评分核对(按评分项、原文定位与分值算术核对评分自洽,仅提示差异,不作出定性结论) | |
383
+ | **[dsh-evidence-check](https://github.com/PerryLink/dsh-evidence-check)** | 证据清单齐备性核对(按待证事实与证据要素核对清单自洽,仅提示差异,不作出定性结论) | |
384
+ | **[dsh-export-ctl-check](https://github.com/PerryLink/dsh-export-ctl-check)** | 出口管制物项核对(按管制清单、最终用途与许可证核对台账齐备,仅提示差异,不作出定性结论) | |
385
+ | **[dsh-fast](https://github.com/PerryLink/dsh-fast)** | Read-only performance diagnostics for DeepSeek Harness. | |
386
+ | **[dsh-fmea-table-check](https://github.com/PerryLink/dsh-fmea-table-check)** | FMEA 分析表要素齐备性与风险顺序数一致性核对(按公开的 FMEA 方法标准核对表内自洽,仅提示差异,不作出定性结论) | |
387
+ | **[dsh-forecast-penalty](https://github.com/PerryLink/dsh-forecast-penalty)** | 预测准确率考核核对(按准确率、门槛与考核单价核对考核费用的算术自洽,仅提示差异,不作出定性结论) | |
388
+ | **[dsh-fund-research](https://github.com/PerryLink/dsh-fund-research)** | Deterministic research reports for Chinese public mutual funds | |
389
+ | **[dsh-github](https://github.com/PerryLink/dsh-github)** | GitHub PR/issues integration for DSH, every write gated by approval | |
390
+ | **[dsh-gongwen-flow-check](https://github.com/PerryLink/dsh-gongwen-flow-check)** | 公文流转与办理时限核对(按公开的公文处理规范核对环节完整性与时限,仅提示差异,不作出定性结论) | |
391
+ | **[dsh-gongwen-word-check](https://github.com/PerryLink/dsh-gongwen-word-check)** | 公文格式要素核对(按公开的党政机关公文格式标准核对版头与主体要素,仅提示差异,不作出定性结论) | |
392
+ | **[dsh-guard-plan-qc](https://github.com/PerryLink/dsh-guard-plan-qc)** | 劳动保护与安全防护用品配置表核对(按公开的配备标准核对品名规格与配置数量,仅提示差异,不作出定性结论) | |
393
+ | **[dsh-hazchem-check](https://github.com/PerryLink/dsh-hazchem-check)** | 危险化学品临界量与重大危险源辨识核对(按储存量与临界量核对台账自洽,仅提示差异,不作出定性结论) | |
394
+ | **[dsh-hazplan-check](https://github.com/PerryLink/dsh-hazplan-check)** | HAZOP 分析表要素齐备性与偏差推导一致性核对(依据公开的危险与可操作性分析标准,仅提示差异,不作出定性结论) | |
395
+ | **[dsh-hidden-risk-map](https://github.com/PerryLink/dsh-hidden-risk-map)** | 安全隐患台账→《工贸企业重大事故隐患判定标准》(应急管理部令第10号)条款映射提示(仅提示差异,不作出定性结论) | |
396
+ | **[dsh-hs-classify](https://github.com/PerryLink/dsh-hs-classify)** | 商品归类层级一致性核对(按章、品目、子目与商品编号的层级关系核对自洽,仅提示差异,不作出定性结论) | |
397
+ | **[dsh-icd-rule-check](https://github.com/PerryLink/dsh-icd-rule-check)** | ICD 合并编码与星剑号配对提示(依据 GB/T 14396-2016 与公开的 ICD 使用说明,仅提示差异,不作出定性结论) | |
398
+ | **[dsh-industry-research](https://github.com/PerryLink/dsh-industry-research)** | Industry research orchestration that seals its deliverables through this plugin's `ctx.researchReport.assemble` | |
399
+ | **[dsh-lawcite-adapter](https://github.com/PerryLink/dsh-lawcite-adapter)** | 法条引用与文书要素核对(按名称、条号、原文与效力状态核对引用可核验性,仅提示差异,不作出定性结论) | |
400
+ | **[dsh-laya](https://github.com/PerryLink/dsh-laya)** | Laya typed decisions (`noul`/`choice`/`score`) as a first-class Cordis service and model-visible tools | |
401
+ | **[dsh-lc-doc-check](https://github.com/PerryLink/dsh-lc-doc-check)** | 信用证单据条款核对(按单据种类、份数、期限与不符点核对交单自洽,仅提示差异,不作出定性结论) | |
402
+ | **[dsh-learning-gap-check](https://github.com/PerryLink/dsh-learning-gap-check)** | 能力差距与培养计划核对(按能力等级差、培养措施与完成期限核对台账闭环,仅提示差异,不作出定性结论) | |
403
+ | **[dsh-library](https://github.com/PerryLink/dsh-library)** | Local document knowledge base for DeepSeek Harness. | |
404
+ | **[dsh-local-ai](https://github.com/PerryLink/dsh-local-ai)** | Local-model (Ollama) integration for DeepSeek Harness. | |
405
+ | **[dsh-lsp-actions](https://github.com/PerryLink/dsh-lsp-actions)** | LSP diagnostics, formatting, completion, code actions and rename over language servers | |
406
+ | **[dsh-mask](https://github.com/PerryLink/dsh-mask)** | PII masking middleware: anonymize at the model boundary, restore at the display layer | |
407
+ | **[dsh-mcp-panel](https://github.com/PerryLink/dsh-mcp-panel)** | Read-only MCP runtime panel: /mcp command + Settings tab with status, tools and errors | |
408
+ | **[dsh-medrec-qc](https://github.com/PerryLink/dsh-medrec-qc)** | 病案首页形式质控与逻辑矛盾提示(依据国卫办医发〔2016〕24号等公开文件,仅提示差异,不作出定性结论) | |
409
+ | **[dsh-memento](https://github.com/PerryLink/dsh-memento)** | Approval-gated cross-session memory: ctx.memory seam + SQLite + memory tool | |
410
+ | **[dsh-nsfc-form-check](https://github.com/PerryLink/dsh-nsfc-form-check)** | 国家自然科学基金申请书形式审查与匿名化提示(依据《国家自然科学基金条例》与申请人导入的当年指南,仅提示差异,不作出定性结论) | |
411
+ | **[dsh-nurse-record-check](https://github.com/PerryLink/dsh-nurse-record-check)** | 护理记录时限、签名与评估表完整性校验(依据《病历书写基本规范》等公开文件,仅提示差异,不作出定性结论) | |
412
+ | **[dsh-observe](https://github.com/PerryLink/dsh-observe)** | OpenTelemetry and Langfuse observability exporter for DeepSeek Harness. | |
413
+ | **[dsh-origin-rvc-check](https://github.com/PerryLink/dsh-origin-rvc-check)** | 原产地区域价值成分核对(按 RVC 定义式与门槛核对原产地判定自洽,仅提示差异,不作出定性结论) | |
414
+ | **[dsh-ota-review-check](https://github.com/PerryLink/dsh-ota-review-check)** | 在线点评回复台账核对(按回复时限、问题分类与整改状态核对台账闭环,仅提示差异,不作出定性结论) | |
415
+ | **[dsh-output-styles](https://github.com/PerryLink/dsh-output-styles)** | Claude Code outputStyles-equivalent runtime style switching | |
416
+ | **[dsh-paper-doc-adapter](https://github.com/PerryLink/dsh-paper-doc-adapter)** | 试卷解析结构核对(按题型、答案、分值与 yotta 标识核对试卷结构自洽,仅提示差异,不作出定性结论) | |
417
+ | **[dsh-permission-rules](https://github.com/PerryLink/dsh-permission-rules)** | Claude Code-style declarative allow/deny/ask permission rules with audit | |
418
+ | **[dsh-permit-report-check](https://github.com/PerryLink/dsh-permit-report-check)** | 行政许可办理核对(按申请材料、法定时限与决定记录核对台账闭环,仅提示差异,不作出定性结论) | |
419
+ | **[dsh-pipeline-check](https://github.com/PerryLink/dsh-pipeline-check)** | 项目全流程文件链条核对(按环节先后与金额口径核对链条自洽,仅提示差异,不作出定性结论) | |
420
+ | **[dsh-pleading-draft](https://github.com/PerryLink/dsh-pleading-draft)** | 起诉状要素齐备性核对(按诉请、事实、证据、法条四要素核对文书齐备,仅提示差异,不作出定性结论) | |
421
+ | **[dsh-plugin-doctor](https://github.com/PerryLink/dsh-plugin-doctor)** | Zero-dependency static + sandbox smoke detector for DSH plugins | |
422
+ | **[dsh-plugin-guide](https://github.com/PerryLink/dsh-plugin-guide)** | Plugin-development knowledge base as an on-demand agent skill | |
423
+ | **[dsh-plugin-kit](https://github.com/PerryLink/dsh-plugin-kit)** | Shared zero-runtime-dependency toolkit for the PerryLink DSH plugins | |
424
+ | **[dsh-plugin-upgrade](https://github.com/PerryLink/dsh-plugin-upgrade)** | One-package, one-corridor-index plugin upgrade skill: routes a repository to the matching closed corridor card | |
425
+ | **[dsh-policy-brief-draft](https://github.com/PerryLink/dsh-policy-brief-draft)** | 政策专报要素核对(按观点、依据、出处与建议核对专报齐备,仅提示差异,不作出定性结论) | |
426
+ | **[dsh-power-loss-split](https://github.com/PerryLink/dsh-power-loss-split)** | 线损电量拆分与考核表核对(依据公开的线损计算口径,仅提示差异,不作出定性结论) | |
427
+ | **[dsh-power-ticket-check](https://github.com/PerryLink/dsh-power-ticket-check)** | 电力工作票要素齐备性与签发许可顺序核对(依据现行电力安全工作规程,仅提示差异,不作出定性结论) | |
428
+ | **[dsh-ppap-check](https://github.com/PerryLink/dsh-ppap-check)** | PPAP 提交要素齐备性核对(按公开的 PPAP 手册与提交等级核对要素齐备与版本填写,仅提示差异,不作出定性结论) | |
429
+ | **[dsh-protest-deadline](https://github.com/PerryLink/dsh-protest-deadline)** | 质疑与投诉期限核对(按公开的政府采购质疑投诉规定核对期限与台账自洽,仅提示差异,不作出定性结论) | |
430
+ | **[dsh-railway-window](https://github.com/PerryLink/dsh-railway-window)** | 铁路施工天窗台账核对(按时间窗、时长算术与命令号核对台账自洽,仅提示差异,不作出定性结论) | |
385
431
  | **[dsh-reach](https://github.com/PerryLink/dsh-reach)** | Multi-channel approval/question bridge: WeChat/Telegram/Feishu, session console | 🧊 FROZEN — see the repo README |
386
- | **[dsh-research-report](https://github.com/PerryLink/dsh-research-report)** | Verifiable research-report engine: content-addressed evidence ledger and sealed versions | |
387
- | **[dsh-score](https://github.com/PerryLink/dsh-score)** | Multi-dimensional quality scoring for DeepSeek Harness plugins. | |
388
- | **[dsh-session-pin](https://github.com/PerryLink/dsh-session-pin)** | Pin sessions in the Web sidebar with durable ordering | 🚫 **RETIRED** — see the note above |
389
- | **[dsh-session-sync](https://github.com/PerryLink/dsh-session-sync)** | Cross-device session sync for DeepSeek Harness — a dedicated git mirror of your session store. | |
390
- | **[dsh-skill-pack-security](https://github.com/PerryLink/dsh-skill-pack-security)** | Security-audit skill pack: secret scan, dependency and supply-chain review | |
391
- | **[dsh-talk](https://github.com/PerryLink/dsh-talk)** | Voice-first session loop for DeepSeek Harness: talk to it, hear it answer. | |
392
- | **[dsh-team-rooms](https://github.com/PerryLink/dsh-team-rooms)** | Cross-session team rooms: shared message bus, task board and timeline | 🚫 **RETIRED** — see the note above |
393
- | **[dsh-test-drive](https://github.com/PerryLink/dsh-test-drive)** | Isolated install-and-smoke test drives for DeepSeek Harness plugins. | |
394
- | **[dsh-ticktick](https://github.com/PerryLink/dsh-ticktick)** | TickTick/Dida365 task bridge: session-header panel + 11 tools | |
395
- | **[dsh-translate](https://github.com/PerryLink/dsh-translate)** | Vendor parameter translation and deterministic JSON repair for DeepSeek Harness. | |
432
+ | **[dsh-repair-order-qc](https://github.com/PerryLink/dsh-repair-order-qc)** | 维修工单结算核对(按项目金额、配件工时合计与结算总额核对算术自洽,仅提示差异,不作出定性结论) | |
433
+ | **[dsh-research-report](https://github.com/PerryLink/dsh-research-report)** | Verifiable research-report engine: content-addressed evidence ledger and sealed versions | |
434
+ | **[dsh-review-reply-check](https://github.com/PerryLink/dsh-review-reply-check)** | 审稿意见逐条回应覆盖核对(按意见与回应一一对应核对覆盖完整性,仅提示差异,不作出定性结论) | |
435
+ | **[dsh-rulefile-check](https://github.com/PerryLink/dsh-rulefile-check)** | 规则库文件自检核对(按条款引用纪律与效力上限核对规则库自身合规,仅提示差异,不作出定性结论) | |
436
+ | **[dsh-safety-brief-check](https://github.com/PerryLink/dsh-safety-brief-check)** | 安全技术交底记录要素齐备性与签字闭环核对(依据公开的安全生产法规与检查标准,仅提示差异,不作出定性结论) | |
437
+ | **[dsh-score](https://github.com/PerryLink/dsh-score)** | Multi-dimensional quality scoring for DeepSeek Harness plugins. | |
438
+ | **[dsh-session-pin](https://github.com/PerryLink/dsh-session-pin)** | Pin sessions in the Web sidebar with durable ordering | 🚫 RETIRED — see the repo README |
439
+ | **[dsh-session-sync](https://github.com/PerryLink/dsh-session-sync)** | Cross-device session sync for DeepSeek Harness — a dedicated git mirror of your session store. | |
440
+ | **[dsh-site-log-check](https://github.com/PerryLink/dsh-site-log-check)** | 监理日志跨日连续性与旁站覆盖提示(依据 GB/T 50319-2013 等公开文件,仅提示差异,不作出定性结论) | |
441
+ | **[dsh-skill-pack-security](https://github.com/PerryLink/dsh-skill-pack-security)** | Security-audit skill pack: secret scan, dependency and supply-chain review | |
442
+ | **[dsh-soe-decision-check](https://github.com/PerryLink/dsh-soe-decision-check)** | 三重一大决策台账核对(按前置研究、会议审议与决议记录核对程序留痕,仅提示差异,不作出定性结论) | |
443
+ | **[dsh-soilwater-check](https://github.com/PerryLink/dsh-soilwater-check)** | 土壤与地下水监测台账核对(按点位、监测项目、限值与超标判定核对台账自洽,仅提示差异,不作出定性结论) | |
444
+ | **[dsh-sop-sync-check](https://github.com/PerryLink/dsh-sop-sync-check)** | 作业规程与 FMEA/控制计划一致性核对(按编号与版本核对三份文件是否同步,仅提示差异,不作出定性结论) | |
445
+ | **[dsh-spc-gbt-adapter](https://github.com/PerryLink/dsh-spc-gbt-adapter)** | 控制图常数与过程能力指数核对(按 GB/T 4091 的常规控制图常数表核对控制限与 Cp/Cpk 的自洽性,仅提示差异,不作出定性结论) | |
446
+ | **[dsh-talk](https://github.com/PerryLink/dsh-talk)** | Voice-first session loop for DeepSeek Harness: talk to it, hear it answer. | |
447
+ | **[dsh-team-rooms](https://github.com/PerryLink/dsh-team-rooms)** | Cross-session team rooms: shared message bus, task board and timeline | 🚫 RETIRED — see the repo README |
448
+ | **[dsh-tender-extract](https://github.com/PerryLink/dsh-tender-extract)** | 招标文件条款摘录核对(按出处与原文逐字核对摘录完整性,仅提示差异,不作出定性结论) | |
449
+ | **[dsh-tender-matrix](https://github.com/PerryLink/dsh-tender-matrix)** | 评标因素评分矩阵核对(按分值上限与合计核对矩阵自洽,仅提示差异,不作出定性结论) | |
450
+ | **[dsh-test-drive](https://github.com/PerryLink/dsh-test-drive)** | Isolated install-and-smoke test drives for DeepSeek Harness plugins. | |
451
+ | **[dsh-ticktick](https://github.com/PerryLink/dsh-ticktick)** | TickTick/Dida365 task bridge: session-header panel + 11 tools | |
452
+ | **[dsh-translate](https://github.com/PerryLink/dsh-translate)** | Vendor parameter translation and deterministic JSON repair for DeepSeek Harness. | |
453
+ | **[dsh-warranty-calc](https://github.com/PerryLink/dsh-warranty-calc)** | 质保期与索赔金额核对(按质保期月数与里程、索赔金额算式核对自洽,仅提示差异,不作出定性结论) | |
454
+
455
+ </details>
456
+
457
+ </details>
396
458
 
397
459
  ### Install from the DSH Desktop Market
398
460
 
@@ -8,6 +8,12 @@
8
8
  * bump the edit count, and a `doublecheck-review` source record marks the
9
9
  * delivery reviewed. Resumed and forked sessions fold to the same view.
10
10
  *
11
+ * The WIRE value is not this state: it is one bounded single-line string this
12
+ * module renders (`renderDoublecheckLine`), because a JSON object of these
13
+ * eight fields is at best 126 characters and every generic one-line renderer
14
+ * truncates it — the old object view lost `gateVerdict`/`gateRedCount`
15
+ * outright. See {@link DOUBLECHECK_LINE_MAX}.
16
+ *
11
17
  * @module dsh-doublecheck/domain/projection
12
18
  */
13
19
  import type { SessionEvent } from '@deepseek-ai/dsh-session';
@@ -15,6 +21,28 @@ import { type TestRunDetection } from './evidence.js';
15
21
  import { type DisciplineStage, type TestColor } from './stages.js';
16
22
  import { type GateVerdict } from './gate.js';
17
23
  import type { DoublecheckView } from '../types.js';
24
+ /**
25
+ * The hard budget of the wire line, in UTF-16 code units — the same unit
26
+ * `String#length` and every one-line renderer measure.
27
+ *
28
+ * 120 is not arbitrary: the WeChat bridge's generic projection renderer
29
+ * (`dsh-wechat` `renderProjectionValue`) keeps a value only while
30
+ * `value.length <= 120` and otherwise cuts it to 117 units plus `…`. The
31
+ * object-shaped view this projection used to emit was 131 units when EMPTY
32
+ * and 139 with a short goal, so it was truncated for every possible value and
33
+ * the two fields a user reads on a phone (`gateVerdict`, `gateRedCount`) were
34
+ * exactly the ones cut off. A single bounded line keeps the whole payload
35
+ * inside every renderer's first line.
36
+ */
37
+ export declare const DOUBLECHECK_LINE_MAX = 120;
38
+ /**
39
+ * Longest `goal:` excerpt kept on the line, in UTF-16 code units. A goal is
40
+ * free text and is the one lossy field: it is sanitized to one line and
41
+ * clipped (never dropped silently — a clipped excerpt ends with `…`). The
42
+ * three fields a user reads first (`gate:`, `stage:`, `r<red>`) are never
43
+ * clipped.
44
+ */
45
+ export declare const DOUBLECHECK_GOAL_MAX = 40;
18
46
  /** The projection unit's plain-JSON fold state. */
19
47
  export interface DoublecheckProjectionState {
20
48
  /** The last discipline stage a successful call or test run reached. */
@@ -58,8 +86,33 @@ export declare function applyDoublecheckEvent(state: DoublecheckProjectionState,
58
86
  */
59
87
  export declare function foldDoublecheckState(events: readonly SessionEvent[], detection: TestRunDetection): DoublecheckProjectionState;
60
88
  /**
61
- * Project the folded state to the wire payload.
89
+ * Project the folded state to the wire payload: one status line.
90
+ *
91
+ * The line is built in priority order and every part is bounded, so the total
92
+ * can never exceed {@link DOUBLECHECK_LINE_MAX}:
93
+ *
94
+ * 1. `gate:<verdict> r<red>` — the latest gate verdict and its red-light count.
95
+ * 2. `stage:<stage>` — the discipline stage reached on durable evidence.
96
+ * 3. `test:<color> edits:<n> spec:<yes|no> review:<yes|no>` — the supporting facts.
97
+ * 4. `goal:<excerpt>` — only when a spec exists and the excerpt is non-empty;
98
+ * clipped to whatever budget parts 1–3 left, at most
99
+ * {@link DOUBLECHECK_GOAL_MAX} units.
100
+ *
101
+ * The worst case of parts 1–3 is 84 units (`gate:deliverable r9999+`,
102
+ * `stage:design`, `test:green edits:99999+ spec:yes review:yes`), leaving at
103
+ * least 28 units for the goal excerpt — so no part is ever dropped to make
104
+ * room, and a final hard clamp backstops the invariant.
62
105
  * @param state - the folded state.
63
106
  * @returns the whole current value for the `doublecheck` projection key.
64
107
  */
65
108
  export declare function viewDoublecheck(state: DoublecheckProjectionState): DoublecheckView;
109
+ /**
110
+ * Render the fold state as the bounded single-line wire value.
111
+ *
112
+ * Pure and total: any state (including a hand-built or version-mismatched one
113
+ * with absurd counts) yields a one-line string of at most
114
+ * {@link DOUBLECHECK_LINE_MAX} code units, never a throw.
115
+ * @param state - the folded state.
116
+ * @returns one line, `<= DOUBLECHECK_LINE_MAX` code units, with no newline.
117
+ */
118
+ export declare function renderDoublecheckLine(state: DoublecheckProjectionState): string;
@@ -8,11 +8,46 @@
8
8
  * bump the edit count, and a `doublecheck-review` source record marks the
9
9
  * delivery reviewed. Resumed and forked sessions fold to the same view.
10
10
  *
11
+ * The WIRE value is not this state: it is one bounded single-line string this
12
+ * module renders (`renderDoublecheckLine`), because a JSON object of these
13
+ * eight fields is at best 126 characters and every generic one-line renderer
14
+ * truncates it — the old object view lost `gateVerdict`/`gateRedCount`
15
+ * outright. See {@link DOUBLECHECK_LINE_MAX}.
16
+ *
11
17
  * @module dsh-doublecheck/domain/projection
12
18
  */
13
19
  import { isTestCommand, isTestFilePath, joinTextBlocks, mutationTargetPath, parseRawArguments, ptcSettle, shellCommand, testOutcome, } from "./evidence.js";
14
20
  import { SPEC_TOOL_NAME } from "./stages.js";
15
21
  import { countRedChecks } from "./gate.js";
22
+ /**
23
+ * The hard budget of the wire line, in UTF-16 code units — the same unit
24
+ * `String#length` and every one-line renderer measure.
25
+ *
26
+ * 120 is not arbitrary: the WeChat bridge's generic projection renderer
27
+ * (`dsh-wechat` `renderProjectionValue`) keeps a value only while
28
+ * `value.length <= 120` and otherwise cuts it to 117 units plus `…`. The
29
+ * object-shaped view this projection used to emit was 131 units when EMPTY
30
+ * and 139 with a short goal, so it was truncated for every possible value and
31
+ * the two fields a user reads on a phone (`gateVerdict`, `gateRedCount`) were
32
+ * exactly the ones cut off. A single bounded line keeps the whole payload
33
+ * inside every renderer's first line.
34
+ */
35
+ export const DOUBLECHECK_LINE_MAX = 120;
36
+ /**
37
+ * Longest `goal:` excerpt kept on the line, in UTF-16 code units. A goal is
38
+ * free text and is the one lossy field: it is sanitized to one line and
39
+ * clipped (never dropped silently — a clipped excerpt ends with `…`). The
40
+ * three fields a user reads first (`gate:`, `stage:`, `r<red>`) are never
41
+ * clipped.
42
+ */
43
+ export const DOUBLECHECK_GOAL_MAX = 40;
44
+ /** Segment separator: visually distinct on a phone and outside the goal word charset. */
45
+ const SEGMENT = ' | ';
46
+ /** Marks a value this line deliberately shortened. */
47
+ const ELLIPSIS = '…';
48
+ /** Above these counts the line prints `<max>+` instead of the exact number. */
49
+ const RED_COUNT_MAX = 9999;
50
+ const EDIT_COUNT_MAX = 99999;
16
51
  /** State for the empty log. */
17
52
  export function emptyDoublecheckState() {
18
53
  return {
@@ -130,19 +165,122 @@ export function foldDoublecheckState(events, detection) {
130
165
  return state;
131
166
  }
132
167
  /**
133
- * Project the folded state to the wire payload.
168
+ * Project the folded state to the wire payload: one status line.
169
+ *
170
+ * The line is built in priority order and every part is bounded, so the total
171
+ * can never exceed {@link DOUBLECHECK_LINE_MAX}:
172
+ *
173
+ * 1. `gate:<verdict> r<red>` — the latest gate verdict and its red-light count.
174
+ * 2. `stage:<stage>` — the discipline stage reached on durable evidence.
175
+ * 3. `test:<color> edits:<n> spec:<yes|no> review:<yes|no>` — the supporting facts.
176
+ * 4. `goal:<excerpt>` — only when a spec exists and the excerpt is non-empty;
177
+ * clipped to whatever budget parts 1–3 left, at most
178
+ * {@link DOUBLECHECK_GOAL_MAX} units.
179
+ *
180
+ * The worst case of parts 1–3 is 84 units (`gate:deliverable r9999+`,
181
+ * `stage:design`, `test:green edits:99999+ spec:yes review:yes`), leaving at
182
+ * least 28 units for the goal excerpt — so no part is ever dropped to make
183
+ * room, and a final hard clamp backstops the invariant.
134
184
  * @param state - the folded state.
135
185
  * @returns the whole current value for the `doublecheck` projection key.
136
186
  */
137
187
  export function viewDoublecheck(state) {
138
- return {
139
- stage: state.stage,
140
- color: state.color,
141
- hasSpec: state.hasSpec,
142
- specGoal: state.specGoal,
143
- reviewed: state.reviewed,
144
- editCount: state.editCount,
145
- gateVerdict: state.gateVerdict,
146
- gateRedCount: state.gateRedCount,
147
- };
188
+ return renderDoublecheckLine(state);
189
+ }
190
+ /**
191
+ * Render the fold state as the bounded single-line wire value.
192
+ *
193
+ * Pure and total: any state (including a hand-built or version-mismatched one
194
+ * with absurd counts) yields a one-line string of at most
195
+ * {@link DOUBLECHECK_LINE_MAX} code units, never a throw.
196
+ * @param state - the folded state.
197
+ * @returns one line, `<= DOUBLECHECK_LINE_MAX` code units, with no newline.
198
+ */
199
+ export function renderDoublecheckLine(state) {
200
+ const head = [
201
+ `gate:${state.gateVerdict} r${boundedCount(state.gateRedCount, RED_COUNT_MAX)}`,
202
+ `stage:${state.stage}`,
203
+ `test:${state.color} edits:${boundedCount(state.editCount, EDIT_COUNT_MAX)}`
204
+ + ` spec:${state.hasSpec ? 'yes' : 'no'} review:${state.reviewed ? 'yes' : 'no'}`,
205
+ ].join(SEGMENT);
206
+ // A goal without a committed spec would contradict `spec:no`.
207
+ const goal = state.hasSpec ? sanitizeGoal(state.specGoal) : '';
208
+ if (goal === '')
209
+ return clampLine(head);
210
+ const budget = DOUBLECHECK_LINE_MAX - head.length - SEGMENT.length - 'goal:'.length;
211
+ if (budget <= 0)
212
+ return clampLine(head);
213
+ const excerpt = capText(goal, Math.min(budget, DOUBLECHECK_GOAL_MAX));
214
+ if (excerpt === '')
215
+ return clampLine(head);
216
+ return clampLine(`${head}${SEGMENT}goal:${excerpt}`);
217
+ }
218
+ /**
219
+ * Render a count for the line: non-finite and negative inputs fold to `0`, and
220
+ * a count above `maximum` prints `<maximum>+` so the field stays bounded.
221
+ * @param value - the folded count.
222
+ * @param maximum - the largest exact count the line prints.
223
+ * @returns a bounded decimal string.
224
+ */
225
+ function boundedCount(value, maximum) {
226
+ if (!Number.isFinite(value))
227
+ return '0';
228
+ const count = Math.max(0, Math.trunc(value));
229
+ return count > maximum ? `${maximum}+` : String(count);
230
+ }
231
+ /**
232
+ * Flatten a free-text goal to one safe line: control characters (including
233
+ * newlines and the C1 range) become spaces, the segment separator becomes `/`,
234
+ * whitespace runs collapse, and the result is trimmed. The wire value must be
235
+ * ONE line — the WeChat renderer keeps only the first line of a multi-line
236
+ * value. `\p{Cc}` is the Unicode control category (U+0000–U+001F and
237
+ * U+007F–U+009F), so no literal control character appears in this pattern.
238
+ * @param goal - the raw committed spec goal.
239
+ * @returns the sanitized single-line goal, possibly empty.
240
+ */
241
+ function sanitizeGoal(goal) {
242
+ return goal
243
+ .replace(/\p{Cc}/gu, ' ')
244
+ .replace(/\|/gu, '/')
245
+ .replace(/\s+/gu, ' ')
246
+ .trim();
247
+ }
248
+ /**
249
+ * Cut `text` to at most `budget` code units, marking a real cut with `…` and
250
+ * never splitting a surrogate pair.
251
+ * @param text - the text to bound.
252
+ * @param budget - the maximum length of the result.
253
+ * @returns the text, or a `…`-terminated prefix of it.
254
+ */
255
+ function capText(text, budget) {
256
+ if (budget <= 0)
257
+ return '';
258
+ if (text.length <= budget)
259
+ return text;
260
+ if (budget <= ELLIPSIS.length)
261
+ return ELLIPSIS;
262
+ return `${text.slice(0, safeEnd(text, budget - ELLIPSIS.length))}${ELLIPSIS}`;
263
+ }
264
+ /**
265
+ * Backstop for {@link DOUBLECHECK_LINE_MAX}: the line is bounded by
266
+ * construction, so this is unreachable today and exists so a future field
267
+ * cannot escape the budget silently.
268
+ * @param line - the assembled line.
269
+ * @returns the line, or a `…`-terminated prefix within budget.
270
+ */
271
+ function clampLine(line) {
272
+ return capText(line, DOUBLECHECK_LINE_MAX);
273
+ }
274
+ /**
275
+ * The largest cut index that does not split a UTF-16 surrogate pair: a lone
276
+ * surrogate is rejected by the WeChat transport (`ret=-1 invalid request`).
277
+ * @param text - the text being cut.
278
+ * @param end - the proposed cut index.
279
+ * @returns `end`, or `end - 1` when it would split a pair.
280
+ */
281
+ function safeEnd(text, end) {
282
+ if (end <= 0)
283
+ return 0;
284
+ const previous = text.charCodeAt(end - 1);
285
+ return previous >= 0xD800 && previous <= 0xDBFF ? end - 1 : end;
148
286
  }
package/lib/types.d.ts CHANGED
@@ -7,39 +7,25 @@
7
7
  * @module dsh-doublecheck/types
8
8
  */
9
9
  import { z as zod } from 'zod';
10
- import type { DisciplineStage, TestColor } from './domain/stages.js';
11
- import type { GateVerdict } from './domain/gate.js';
12
10
  import type { DoublecheckProjectionState } from './domain/projection.js';
13
- /** The whole wire value of the `doublecheck` session projection. */
14
- export interface DoublecheckView {
15
- /** The last discipline stage reached on durable evidence. */
16
- stage: DisciplineStage;
17
- /** The latest test-run evidence color. */
18
- color: TestColor;
19
- /** Whether a committed requirements spec exists in the log. */
20
- hasSpec: boolean;
21
- /** The goal of the latest committed spec, or '' before any spec. */
22
- specGoal: string;
23
- /** Whether an adversary review record exists in the log. */
24
- reviewed: boolean;
25
- /** Total implementation edits folded so far. */
26
- editCount: number;
27
- /** The latest delivery-gate verdict, or 'none' before any gate run. */
28
- gateVerdict: 'none' | GateVerdict;
29
- /** Red-light (failing) checks of the latest gate run. */
30
- gateRedCount: number;
31
- }
11
+ /**
12
+ * The whole wire value of the `doublecheck` session projection: ONE bounded
13
+ * status line, e.g.
14
+ *
15
+ * ```text
16
+ * gate:rework r2 | stage:green | test:green edits:12 spec:yes review:yes | goal:ship parser fix
17
+ * ```
18
+ *
19
+ * It is a string rather than an object on purpose. The eight state fields cost
20
+ * 126 characters as compact JSON even at their shortest possible values, so no
21
+ * object-shaped payload can survive a renderer that keeps one line under 121
22
+ * characters — which is exactly how `gateVerdict` and `gateRedCount` used to
23
+ * disappear. `renderDoublecheckLine` (`./domain/projection.ts`) owns the
24
+ * format and the {@link DOUBLECHECK_LINE_MAX} budget.
25
+ */
26
+ export type DoublecheckView = string;
32
27
  /** Validates the `doublecheck` projection's wire payload before it leaves the host. */
33
- export declare const doublecheckViewSchema: zod.ZodObject<{
34
- stage: zod.ZodUnion<readonly [zod.ZodLiteral<"grill">, zod.ZodLiteral<"design">, zod.ZodLiteral<"red">, zod.ZodLiteral<"green">, zod.ZodLiteral<"review">, zod.ZodLiteral<"verify">]>;
35
- color: zod.ZodUnion<readonly [zod.ZodLiteral<"none">, zod.ZodLiteral<"red">, zod.ZodLiteral<"green">]>;
36
- hasSpec: zod.ZodBoolean;
37
- specGoal: zod.ZodString;
38
- reviewed: zod.ZodBoolean;
39
- editCount: zod.ZodNumber;
40
- gateVerdict: zod.ZodUnion<readonly [zod.ZodLiteral<"none">, zod.ZodLiteral<"deliverable">, zod.ZodLiteral<"rework">]>;
41
- gateRedCount: zod.ZodNumber;
42
- }, zod.core.$strip>;
28
+ export declare const doublecheckViewSchema: zod.ZodString;
43
29
  /** Validates the persisted `doublecheck` projection state before it seeds a fold. */
44
30
  export declare const doublecheckStateSchema: zod.ZodObject<{
45
31
  stage: zod.ZodUnion<readonly [zod.ZodLiteral<"grill">, zod.ZodLiteral<"design">, zod.ZodLiteral<"red">, zod.ZodLiteral<"green">, zod.ZodLiteral<"review">, zod.ZodLiteral<"verify">]>;
package/lib/types.js CHANGED
@@ -7,17 +7,11 @@
7
7
  * @module dsh-doublecheck/types
8
8
  */
9
9
  import { z as zod } from 'zod';
10
+ // Value import (the one runtime edge in this type table): the wire line's
11
+ // budget is declared once, next to the renderer that must honor it.
12
+ import { DOUBLECHECK_LINE_MAX } from "./domain/projection.js";
10
13
  /** Validates the `doublecheck` projection's wire payload before it leaves the host. */
11
- export const doublecheckViewSchema = zod.object({
12
- stage: zod.union([zod.literal('grill'), zod.literal('design'), zod.literal('red'), zod.literal('green'), zod.literal('review'), zod.literal('verify')]),
13
- color: zod.union([zod.literal('none'), zod.literal('red'), zod.literal('green')]),
14
- hasSpec: zod.boolean(),
15
- specGoal: zod.string(),
16
- reviewed: zod.boolean(),
17
- editCount: zod.number(),
18
- gateVerdict: zod.union([zod.literal('none'), zod.literal('deliverable'), zod.literal('rework')]),
19
- gateRedCount: zod.number(),
20
- });
14
+ export const doublecheckViewSchema = zod.string().max(DOUBLECHECK_LINE_MAX);
21
15
  /** Validates the persisted `doublecheck` projection state before it seeds a fold. */
22
16
  export const doublecheckStateSchema = zod.object({
23
17
  stage: zod.union([zod.literal('grill'), zod.literal('design'), zod.literal('red'), zod.literal('green'), zod.literal('review'), zod.literal('verify')]),
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "dsh-doublecheck",
3
- "version": "0.9.26",
3
+ "version": "0.9.27",
4
4
  "description": "Delivery quality gate for DeepSeek Harness: grill the requirements, test the implementation, prove the delivery, and gate the handoff with a deliverable/rework decision.",
5
5
  "packageManager": "pnpm@11.7.0",
6
6
  "repository": {