universal-dev-standards 6.7.5 → 6.9.0
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/bin/uds.js +19 -2
- package/bundled/ai/standards/acceptance-criteria-traceability.ai.yaml +14 -2
- package/bundled/ai/standards/adr-standards.ai.yaml +14 -2
- package/bundled/ai/standards/ai-instruction-standards.ai.yaml +6 -6
- package/bundled/ai/standards/code-review.ai.yaml +13 -3
- package/bundled/ai/standards/commit-message.ai.yaml +8 -4
- package/bundled/ai/standards/deferred-item-exit.ai.yaml +225 -0
- package/bundled/ai/standards/feature-discovery-standards.ai.yaml +14 -2
- package/bundled/ai/standards/governance-layer.ai.yaml +128 -2
- package/bundled/ai/standards/logging.ai.yaml +2 -2
- package/bundled/ai/standards/retrospective-standards.ai.yaml +14 -2
- package/bundled/ai/standards/reverse-engineering-standards.ai.yaml +73 -2
- package/bundled/ai/standards/security-standards.ai.yaml +2 -2
- package/bundled/ai/standards/spec-driven-development.ai.yaml +14 -2
- package/bundled/ai/standards/tech-debt-standards.ai.yaml +87 -3
- package/bundled/ai/standards/turn-completion-integrity.ai.yaml +131 -0
- package/bundled/core/acceptance-criteria-traceability.md +5 -2
- package/bundled/core/adr-standards.md +26 -2
- package/bundled/core/ai-instruction-standards.md +9 -7
- package/bundled/core/code-review-checklist.md +5 -2
- package/bundled/core/context-aware-loading.md +1 -1
- package/bundled/core/deferred-item-exit.md +254 -0
- package/bundled/core/feature-discovery-standards.md +5 -1
- package/bundled/core/governance-layer.md +114 -2
- package/bundled/core/retrospective-standards.md +4 -2
- package/bundled/core/reverse-engineering-standards.md +81 -2
- package/bundled/core/spec-driven-development.md +8 -2
- package/bundled/core/tech-debt-standards.md +67 -8
- package/bundled/core/turn-completion-integrity.md +196 -0
- package/bundled/hooks/check-dangerous-cmd.mjs +60 -0
- package/bundled/hooks/check-logging-standard.mjs +59 -0
- package/bundled/hooks/check-turn-completion.mjs +233 -0
- package/bundled/hooks/inject-standards.mjs +183 -0
- package/bundled/hooks/telemetry-wrapper.mjs +77 -0
- package/bundled/hooks/turn-completion/detect.mjs +99 -0
- package/bundled/hooks/turn-completion/locales/en.mjs +159 -0
- package/bundled/hooks/turn-completion/locales/zh-TW.mjs +166 -0
- package/bundled/hooks/validate-commit-msg.mjs +104 -0
- package/bundled/locales/zh-CN/CHANGELOG.md +65 -3
- package/bundled/locales/zh-CN/CLAUDE.md +1 -1
- package/bundled/locales/zh-CN/README.md +7 -7
- package/bundled/locales/zh-CN/SECURITY.md +1 -1
- package/bundled/locales/zh-CN/core/adr-standards.md +1 -1
- package/bundled/locales/zh-CN/core/ai-instruction-standards.md +10 -8
- package/bundled/locales/zh-CN/core/governance-layer.md +118 -6
- package/bundled/locales/zh-CN/core/retrospective-standards.md +1 -1
- package/bundled/locales/zh-CN/core/tech-debt-standards.md +71 -4
- package/bundled/locales/zh-CN/core/turn-completion-integrity.md +190 -0
- package/bundled/locales/zh-CN/docs/CHEATSHEET.md +27 -6
- package/bundled/locales/zh-CN/docs/CLI-INIT-OPTIONS.md +29 -68
- package/bundled/locales/zh-CN/docs/FEATURE-REFERENCE.md +172 -24
- package/bundled/locales/zh-CN/docs/USAGE-MODES-COMPARISON.md +1 -2
- package/bundled/locales/zh-CN/integrations/google-antigravity/{INSTRUCTIONS.md → AGENTS.md} +1 -1
- package/bundled/locales/zh-CN/integrations/google-antigravity/README.md +3 -3
- package/bundled/locales/zh-CN/skills/agents/README.md +1 -1
- package/bundled/locales/zh-CN/skills/atdd-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-CN/skills/bdd-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-CN/skills/brainstorm-assistant/SKILL.md +22 -12
- package/bundled/locales/zh-CN/skills/brainstorm-assistant/guide.md +12 -9
- package/bundled/locales/zh-CN/skills/code-review-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-CN/skills/commands/brainstorm.md +17 -13
- package/bundled/locales/zh-CN/skills/commands/config.md +0 -1
- package/bundled/locales/zh-CN/skills/commands/init.md +1 -2
- package/bundled/locales/zh-CN/skills/commit-standards/SKILL.md +2 -0
- package/bundled/locales/zh-CN/skills/contract-test-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-CN/skills/dev-methodology/SKILL.md +2 -0
- package/bundled/locales/zh-CN/skills/observability-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-CN/skills/project-structure-guide/SKILL.md +1 -0
- package/bundled/locales/zh-CN/skills/release-standards/SKILL.md +3 -0
- package/bundled/locales/zh-CN/skills/requirement-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-CN/skills/reverse-engineer/SKILL.md +3 -0
- package/bundled/locales/zh-CN/skills/reverse-engineer/tdd-analysis.md +13 -23
- package/bundled/locales/zh-CN/skills/runbook-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-CN/skills/slo-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-CN/skills/tdd-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-CN/skills/workflows/README.md +2 -11
- package/bundled/locales/zh-TW/CHANGELOG.md +65 -3
- package/bundled/locales/zh-TW/CLAUDE.md +1 -1
- package/bundled/locales/zh-TW/README.md +7 -7
- package/bundled/locales/zh-TW/SECURITY.md +1 -1
- package/bundled/locales/zh-TW/core/acceptance-criteria-traceability.md +2 -0
- package/bundled/locales/zh-TW/core/adr-standards.md +26 -5
- package/bundled/locales/zh-TW/core/ai-instruction-standards.md +10 -8
- package/bundled/locales/zh-TW/core/code-review-checklist.md +2 -0
- package/bundled/locales/zh-TW/core/container-image-standards.md +2 -2
- package/bundled/locales/zh-TW/core/contract-testing-standards.md +2 -2
- package/bundled/locales/zh-TW/core/cross-flow-regression.md +8 -7
- package/bundled/locales/zh-TW/core/data-contract.md +2 -2
- package/bundled/locales/zh-TW/core/data-migration-testing.md +2 -2
- package/bundled/locales/zh-TW/core/data-pipeline.md +2 -2
- package/bundled/locales/zh-TW/core/deferred-item-exit.md +251 -0
- package/bundled/locales/zh-TW/core/documentation-writing-standards.md +228 -3
- package/bundled/locales/zh-TW/core/full-coverage-testing.md +15 -2
- package/bundled/locales/zh-TW/core/governance-layer.md +118 -5
- package/bundled/locales/zh-TW/core/iac-design-principles.md +2 -2
- package/bundled/locales/zh-TW/core/incident-response.md +2 -2
- package/bundled/locales/zh-TW/core/model-provenance.md +4 -2
- package/bundled/locales/zh-TW/core/pii-classification.md +42 -6
- package/bundled/locales/zh-TW/core/prd-standards.md +4 -2
- package/bundled/locales/zh-TW/core/product-metrics-standards.md +4 -2
- package/bundled/locales/zh-TW/core/release-readiness-gate.md +2 -2
- package/bundled/locales/zh-TW/core/resource-cost-boundary.md +2 -2
- package/bundled/locales/zh-TW/core/retrospective-standards.md +5 -3
- package/bundled/locales/zh-TW/core/reverse-engineering-standards.md +83 -5
- package/bundled/locales/zh-TW/core/runbook.md +2 -2
- package/bundled/locales/zh-TW/core/schema-evolution.md +2 -2
- package/bundled/locales/zh-TW/core/secret-management-standards.md +2 -2
- package/bundled/locales/zh-TW/core/slo-sli.md +2 -2
- package/bundled/locales/zh-TW/core/spec-driven-development.md +2 -0
- package/bundled/locales/zh-TW/core/tech-debt-standards.md +71 -4
- package/bundled/locales/zh-TW/core/turn-completion-integrity.md +190 -0
- package/bundled/locales/zh-TW/core/user-journey-testing.md +2 -2
- package/bundled/locales/zh-TW/core/user-story-mapping.md +2 -2
- package/bundled/locales/zh-TW/core/verification-oracle.md +2 -2
- package/bundled/locales/zh-TW/docs/CHEATSHEET.md +27 -6
- package/bundled/locales/zh-TW/docs/CLI-INIT-OPTIONS.md +29 -68
- package/bundled/locales/zh-TW/docs/FEATURE-REFERENCE.md +172 -24
- package/bundled/locales/zh-TW/docs/USAGE-MODES-COMPARISON.md +1 -2
- package/bundled/locales/zh-TW/integrations/google-antigravity/{INSTRUCTIONS.md → AGENTS.md} +1 -1
- package/bundled/locales/zh-TW/integrations/google-antigravity/README.md +3 -3
- package/bundled/locales/zh-TW/skills/adr-assistant/SKILL.md +1 -1
- package/bundled/locales/zh-TW/skills/agents/README.md +1 -1
- package/bundled/locales/zh-TW/skills/atdd-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-TW/skills/bdd-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-TW/skills/brainstorm-assistant/SKILL.md +22 -12
- package/bundled/locales/zh-TW/skills/brainstorm-assistant/guide.md +12 -9
- package/bundled/locales/zh-TW/skills/code-review-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-TW/skills/commands/brainstorm.md +17 -13
- package/bundled/locales/zh-TW/skills/commands/config.md +0 -1
- package/bundled/locales/zh-TW/skills/commands/init.md +1 -2
- package/bundled/locales/zh-TW/skills/commit-standards/SKILL.md +2 -0
- package/bundled/locales/zh-TW/skills/contract-test-assistant/SKILL.md +2 -1
- package/bundled/locales/zh-TW/skills/dev-methodology/SKILL.md +2 -0
- package/bundled/locales/zh-TW/skills/dev-workflow-guide/SKILL.md +1 -1
- package/bundled/locales/zh-TW/skills/knowledge-graph/guide.md +2 -2
- package/bundled/locales/zh-TW/skills/migration-assistant/SKILL.md +1 -1
- package/bundled/locales/zh-TW/skills/observability-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-TW/skills/project-discovery/SKILL.md +1 -0
- package/bundled/locales/zh-TW/skills/project-structure-guide/SKILL.md +1 -0
- package/bundled/locales/zh-TW/skills/release-standards/SKILL.md +3 -0
- package/bundled/locales/zh-TW/skills/requirement-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-TW/skills/reverse-engineer/SKILL.md +3 -0
- package/bundled/locales/zh-TW/skills/reverse-engineer/tdd-analysis.md +13 -23
- package/bundled/locales/zh-TW/skills/runbook-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-TW/skills/slo-assistant/SKILL.md +1 -0
- package/bundled/locales/zh-TW/skills/tdd-assistant/SKILL.md +2 -0
- package/bundled/locales/zh-TW/skills/workflows/README.md +2 -11
- package/bundled/skills/agents/README.md +1 -1
- package/bundled/skills/atdd-assistant/SKILL.md +2 -0
- package/bundled/skills/bdd-assistant/SKILL.md +2 -0
- package/bundled/skills/brainstorm-assistant/SKILL.md +31 -13
- package/bundled/skills/brainstorm-assistant/guide.md +9 -6
- package/bundled/skills/code-review-assistant/SKILL.md +1 -0
- package/bundled/skills/commands/brainstorm.md +12 -9
- package/bundled/skills/commands/config.md +0 -1
- package/bundled/skills/commands/init.md +2 -3
- package/bundled/skills/commit-standards/SKILL.md +2 -0
- package/bundled/skills/contract-test-assistant/SKILL.md +1 -0
- package/bundled/skills/dev-methodology/SKILL.md +4 -0
- package/bundled/skills/observability-assistant/SKILL.md +1 -0
- package/bundled/skills/project-discovery/SKILL.md +1 -0
- package/bundled/skills/project-structure-guide/SKILL.md +1 -0
- package/bundled/skills/release-standards/SKILL.md +3 -0
- package/bundled/skills/requirement-assistant/SKILL.md +2 -0
- package/bundled/skills/reverse-engineer/SKILL.md +3 -0
- package/bundled/skills/reverse-engineer/tdd-analysis.md +16 -23
- package/bundled/skills/runbook-assistant/SKILL.md +1 -0
- package/bundled/skills/slo-assistant/SKILL.md +1 -0
- package/bundled/skills/tdd-assistant/SKILL.md +2 -0
- package/bundled/skills/workflows/README.md +2 -11
- package/bundled/templates/.ai-context.yaml.template +194 -0
- package/bundled/templates/CLAUDE.md.template +145 -0
- package/bundled/templates/DESIGN.md +237 -0
- package/bundled/templates/SKILL-BRIEF-TEMPLATE.md +57 -0
- package/bundled/templates/SKILL-CANDIDATES.md +39 -0
- package/bundled/templates/gates/check-error-exit.mjs +309 -0
- package/bundled/templates/mcp-config.json +10 -0
- package/bundled/templates/methodology-template.yaml +209 -0
- package/bundled/templates/migration-template.md +408 -0
- package/bundled/templates/requirement-checklist.md +410 -0
- package/bundled/templates/requirement-document-template.md +591 -0
- package/bundled/templates/requirement-template.md +881 -0
- package/bundled/templates/reverse-spec-template.md +409 -0
- package/bundled/templates/test-case-template.md +74 -0
- package/bundled/templates/test-plan-template.md +74 -0
- package/package.json +9 -5
- package/src/commands/audit.js +82 -0
- package/src/commands/check.js +66 -10
- package/src/commands/init.js +161 -16
- package/src/commands/lint.js +96 -0
- package/src/commands/quickstart.js +16 -13
- package/src/commands/update.js +286 -14
- package/src/compilers/claude-code-compiler.js +4 -1
- package/src/config/ai-agent-paths.js +62 -17
- package/src/core/constants.js +42 -11
- package/src/core/manifest.js +201 -3
- package/src/core/paths.js +2 -2
- package/src/i18n/messages.js +9 -32
- package/src/installers/hooks-installer.js +167 -75
- package/src/installers/integration-installer.js +9 -5
- package/src/prompts/init.js +14 -14
- package/src/reconciler/actual-state-scanner.js +14 -3
- package/src/utils/detector.js +21 -1
- package/src/utils/effect-boundary.js +1093 -0
- package/src/utils/hasher.js +229 -9
- package/src/utils/hook-stats.js +1 -1
- package/src/utils/integration-generator.js +100 -4
- package/src/utils/reference-sync.js +4 -1
- package/src/utils/skills-installer.js +17 -3
- package/src/utils/spec-linter.js +35 -76
- package/src/utils/yaml-generator.js +51 -9
- package/standards-registry.json +31 -8
- package/src/commands/sync.js +0 -133
|
@@ -5,10 +5,10 @@ description: |
|
|
|
5
5
|
使用時機:模糊想法、功能探索、問題再框架、創意發想。
|
|
6
6
|
關鍵字:brainstorm, ideation, persona ensemble, multi-critic, HMW, SCAMPER, 腦力激盪, 發想, 創意。
|
|
7
7
|
source: ../../../../skills/brainstorm-assistant/guide.md
|
|
8
|
-
source_version: 4.
|
|
9
|
-
source_hash:
|
|
10
|
-
translation_version: 4.
|
|
11
|
-
last_synced: 2026-
|
|
8
|
+
source_version: 4.2.0
|
|
9
|
+
source_hash: f789b3bbc31a
|
|
10
|
+
translation_version: 4.2.0
|
|
11
|
+
last_synced: 2026-08-21
|
|
12
12
|
status: current
|
|
13
13
|
---
|
|
14
14
|
|
|
@@ -16,8 +16,8 @@ status: current
|
|
|
16
16
|
|
|
17
17
|
> **語言**: [English](../../../../skills/brainstorm-assistant/guide.md) | 繁體中文
|
|
18
18
|
|
|
19
|
-
**版本**: 4.
|
|
20
|
-
**最後更新**: 2026-
|
|
19
|
+
**版本**: 4.2.0
|
|
20
|
+
**最後更新**: 2026-08-21
|
|
21
21
|
**適用範圍**: 所有軟體專案
|
|
22
22
|
**Scope**: universal
|
|
23
23
|
**類型**: 工具型 Skill(無對應核心標準)
|
|
@@ -280,7 +280,7 @@ HMW(預設起點)、SCAMPER(改善既有功能:Substitute 取代、Combi
|
|
|
280
280
|
| **使用者衝擊** | 衝擊 70% · 對齊 30% |
|
|
281
281
|
| **策略對齊** | 對齊 60% · 衝擊 40% |
|
|
282
282
|
|
|
283
|
-
**各準則指引(1–5
|
|
283
|
+
**各準則指引(1–5,參考格式——不是評審必須逐步照走的腳本):**
|
|
284
284
|
|
|
285
285
|
| 準則 | 5 | 3 | 1 |
|
|
286
286
|
|------|---|---|---|
|
|
@@ -306,7 +306,7 @@ HMW(預設起點)、SCAMPER(改善既有功能:Substitute 取代、Combi
|
|
|
306
306
|
- **Devil's Advocate(魔鬼代言人)**:「你的任務是論證這個想法*會*失敗。提出 2 個具體的失敗條件。」
|
|
307
307
|
- **Steelman(強論立場)**:「陳述最強、最善意版本的反論點——一個有思考力的對手真的會提出的那種。」
|
|
308
308
|
|
|
309
|
-
|
|
309
|
+
約束:每個反論點都必須指出**特定脈絡**與**特定理由**。句型**「這個想法會在 [特定脈絡] 失敗,因為 [特定理由]」**是參考格式,不是必用句式(v4.2——評審提示為目標+約束,見 SKILL.md「評審提示形狀」)。
|
|
310
310
|
|
|
311
311
|
**不可接受**(太空泛):「這可能很難。」/「可能會有 edge case。」
|
|
312
312
|
|
|
@@ -342,6 +342,7 @@ HMW(預設起點)、SCAMPER(改善既有功能:Substitute 取代、Combi
|
|
|
342
342
|
**Personas Used**: [領域專家, 懷疑者, 類比者, ...]
|
|
343
343
|
**Lenses Used**: [類比, 反轉, ...]
|
|
344
344
|
**Pre-flight**: [Completed / Skipped] **Rebuttal**: [Completed / Skipped] **Tier**: [Baseline / Enhanced]
|
|
345
|
+
**評審模型層級(D4 心智軸)**: [core/model-selection.md 的廠商中立層級標籤——指名廠商模型即違規;未宣告 → D4 [degraded]]
|
|
345
346
|
|
|
346
347
|
## Problem Statement
|
|
347
348
|
[精煉後的問題 + 5 Whys 的根本原因]
|
|
@@ -549,7 +550,7 @@ BQS v1 是**疊加的品質契約**,不是打掉重練。v3 的一切都保留
|
|
|
549
550
|
|
|
550
551
|
- **所有旗標**:`--personas`、`--lens`、`--enhanced`、`--skip-preflight`、`--no-rebuttal`、`--quick`、`--technique` 行為與 v3 完全相同。v4 僅新增一個可選旗標 `--intent`,供第 0 層宣告。
|
|
551
552
|
- **所有機制**:PRE-FLIGHT 防錨定、FRAME 5-Whys/HMW、persona 集成、多樣性透鏡、多評審面板、硬角色反駁(Devil's Advocate + Steelman)、多樣性崩塌護欄、Enhanced 層,皆不變。
|
|
552
|
-
- **`--enhanced` 隔離 agent
|
|
553
|
+
- **`--enhanced` 隔離 agent 宿主**滿足的是 BQS **D4 的座位軸**(判官≠產生者);自 v4.2(XSPEC-388)起,D4 pass 另需評審宣告**模型層級**(心智軸——依 core/model-selection.md 的廠商中立標籤)。baseline 單 context 標 `[degraded]`、不靜默通過。
|
|
553
554
|
|
|
554
555
|
---
|
|
555
556
|
|
|
@@ -584,6 +585,7 @@ BQS v1 是**疊加的品質契約**,不是打掉重練。v3 的一切都保留
|
|
|
584
585
|
| [Requirement Engineering](../../core/requirement-engineering.md) | 腦力激盪輸出餵入需求撰寫 |
|
|
585
586
|
| [Spec-Driven Development](../../core/spec-driven-development.md) | 腦力激盪輸出餵入 SDD 提案 |
|
|
586
587
|
| [Anti-Hallucination](../../core/anti-hallucination.md) | 評審的可行性宣稱必須以證據為基礎,不可空口斷言 |
|
|
588
|
+
| [Model Selection](../../core/model-selection.md) | D4 心智軸(宣告模型層級)與發散側層級(Standard 或更高)皆以此標準的判準為準——引用、絕不複述 |
|
|
587
589
|
|
|
588
590
|
---
|
|
589
591
|
|
|
@@ -591,6 +593,7 @@ BQS v1 是**疊加的品質契約**,不是打掉重練。v3 的一切都保留
|
|
|
591
593
|
|
|
592
594
|
| 版本 | 日期 | 變更 |
|
|
593
595
|
|------|------|------|
|
|
596
|
+
| 4.2.0 | 2026-08-21 | XSPEC-388:D4 的 pass 條件加上第二個軸——座位(獨立 context,不變)加上心智(評審必須宣告其**模型層級**要求,且只能以 core/model-selection.md 的廠商中立標籤表達;指名具體廠商模型即違規;未宣告 → `[degraded]`、不得 pass)。CONVERGE 評審提示依 MS-009 改為目標+約束形狀:D5–D8 以「判斷必須滿足的判準」呈現、不得以按順序執行的程序呈現;固定反駁句型、逐準則 1–5 指南與評審權重公式保留為參考格式、非必經步驟。發散側引用 MS-004/Criterion 2(Standard 或更高——刻意不是「最高」),由校準回路既有的 Diversity(D2/D3)lagging 欄驗證。術語:模型能力軸全文寫作「模型層級」,因為「tier」已被 BQS 分級與 Enhanced Tier 佔用。所有 v3 旗標與機制保留。 |
|
|
594
597
|
| 4.1.0 | 2026-07-09 | XSPEC-325:把固定 Top-3 推薦上限改為分數門檻**推薦集**(Agg. Score ≥ 3.5,不限筆數),貫穿 BQS 第 2 層——產物閘(D5–D8)、硬角色反駁輪(Devil's Advocate + Steelman)、Meta 停止規則(集合成員穩定性)、多樣性崩塌護欄、OUTPUT 推薦區塊,全部改以推薦集為對象,不再是固定 3 個。若無想法達 3.5,仍顯示分數最高的 1 個並標示「未達門檻,僅供參考」,避免報告空白。 |
|
|
595
598
|
| 4.0.0 | 2026-06-22 | XSPEC-296:腦力激盪品質標準(BQS v1)——疊加於 v3 的四層 × 時間軸品質契約。第 0 層 explore/exploit 意圖(調節 D2 權重);第 1 層過程 leading 維度 D1–D4 + 硬序列閘(發散期禁 D5–D8);第 2 層產物 leading 維度 D5–D8 僅施於 Top 3,加 Seeds 欄與高方差爭議區;第 3 層 Judgment Override(凌駕聚合分)。D4 判官≠產生者(須獨立 context,否則 `[degraded]`);D5 主張分流(外部事實跨級地板);D7 二態證偽(「需先做 X」→next-step、不算 fail);Meta 停止規則(Top-3 集合穩定 + 硬上限 2 輪)。工作階段自評重新定位為校準回路 lagging 端(Adoption→D6、Diversity→D2/D3、Cognitive Load→成本;禁兩套平行評估)。第一原理修正為「決策用 leading、校準用 lagging」。分級綁客觀模式選擇觸發。新增旗標 `--intent`。所有 v3 旗標/機制保留。 |
|
|
596
599
|
| 3.0.0 | 2026-06-01 | XSPEC-247:DIVERGE 重新對齊到 persona 集成 + 多樣性透鏡(類比/反轉/形態學);CONVERGE 重新對齊到多評審面板 + 硬角色反駁(Devil's Advocate + Steelman);多樣性崩塌護欄;Enhanced 層級(平行 persona/評審代理、優雅退回);認知科學依據以 6 項已驗證的 2024–2026 出處重建;效度說明重新評等(pre-flight 為低、Nijstad/Nemeth 降級);新增旗標 `--personas`/`--lens`/`--enhanced`;反種子護欄 |
|
|
@@ -1,9 +1,10 @@
|
|
|
1
1
|
---
|
|
2
2
|
source: ../../../../skills/commands/brainstorm.md
|
|
3
|
-
source_version: 4.
|
|
4
|
-
source_hash:
|
|
5
|
-
translation_version: 4.
|
|
6
|
-
last_synced: 2026-
|
|
3
|
+
source_version: 4.2.0
|
|
4
|
+
source_hash: 319f8ad4f02f
|
|
5
|
+
translation_version: 4.2.0
|
|
6
|
+
last_synced: 2026-08-21
|
|
7
|
+
status: current
|
|
7
8
|
---
|
|
8
9
|
|
|
9
10
|
---
|
|
@@ -36,7 +37,7 @@ PRE-FLIGHT ──► FRAME ──► DIVERGE ──────────► C
|
|
|
36
37
|
| **PRE-FLIGHT** | 防止 AI 錨定 | 使用者先寫 3 個想法;無類比種子;**BQS 第 0 層意圖(explore/exploit)** |
|
|
37
38
|
| **FRAME** | 清楚定義問題 | 5 Whys、HMW、Stakeholder Map;**D1 框架純度** |
|
|
38
39
|
| **DIVERGE** | 逼出視角多樣性 | persona 集成 + 多樣性透鏡;**BQS 第 1 層 D1–D4,硬閘隱藏 D5–D8** |
|
|
39
|
-
| **CONVERGE** | 降偏誤選擇 | 多評審面板 + Devil's Advocate/Steelman;**BQS 第 2 層 D5–D8
|
|
40
|
+
| **CONVERGE** | 降偏誤選擇 | 多評審面板 + Devil's Advocate/Steelman;**BQS 第 2 層 D5–D8 施於推薦集(Agg. ≥ 3.5),D4 判官≠產生者(兩軸:獨立 context+宣告模型層級)** |
|
|
40
41
|
| **OUTPUT** | 可執行的報告 | Brainstorm Report + Seeds + 爭議區 + **第 3 層 Judgment Override** |
|
|
41
42
|
|
|
42
43
|
> **BQS v1(XSPEC-296):** v4 在 v3 上疊加四層 × 時間軸品質契約。完整 oracle 與結構規則見 skill:[Brainstorm Assistant Skill](../brainstorm-assistant/SKILL.md#bqs-v1--品質契約)。
|
|
@@ -70,7 +71,7 @@ PRE-FLIGHT ──► FRAME ──► DIVERGE ──────────► C
|
|
|
70
71
|
## Problem Statement
|
|
71
72
|
[Refined problem from FRAME phase]
|
|
72
73
|
|
|
73
|
-
##
|
|
74
|
+
## Recommendations (Agg. Score ≥ 3.5 — uncapped)
|
|
74
75
|
1. **[Idea]** — Agg. X.X ✓ Passed rebuttal — Persona: [..] — [Why recommended]
|
|
75
76
|
|
|
76
77
|
## Diversity Note
|
|
@@ -102,7 +103,7 @@ PRE-FLIGHT ──► FRAME ──► DIVERGE ──────────► C
|
|
|
102
103
|
| `/brainstorm "topic"` | 以指定主題啟動 PRE-FLIGHT,再進入 FRAME |
|
|
103
104
|
| `/brainstorm --personas "a,b,c"` | 以自訂 persona 組進入 DIVERGE |
|
|
104
105
|
| `/brainstorm --lens <name>` | 以指定多樣性透鏡為主進入 DIVERGE |
|
|
105
|
-
| `/brainstorm --enhanced` | 若宿主支援子代理則平行跑 persona/評審,否則靜默退回 baseline
|
|
106
|
+
| `/brainstorm --enhanced` | 若宿主支援子代理則平行跑 persona/評審,否則靜默退回 baseline;此模式滿足 BQS D4 的座位軸(獨立 context),pass 另需宣告模型層級(心智軸,見 skill D4) |
|
|
106
107
|
| `/brainstorm --intent exploit` | 以 exploit 意圖啟動(BQS 第 0 層),D2 低覆蓋不扣分 |
|
|
107
108
|
|
|
108
109
|
### 互動腳本
|
|
@@ -127,12 +128,13 @@ PRE-FLIGHT ──► FRAME ──► DIVERGE ──────────► C
|
|
|
127
128
|
3. 至少套用一個多樣性透鏡(類比/假設反轉/形態矩陣)
|
|
128
129
|
4. 以多樣性(覆蓋的 persona/透鏡數,D2/D3)而非數量為繼續門檻
|
|
129
130
|
5. **硬序列閘**:D5–D8 在最後一個 persona 產完前**不得揭示或評分**;CONVERGE critic 不得提前呼叫
|
|
131
|
+
6. **模型層級(發散側)**:以 **Standard 或更高**層級執行——依 core/model-selection.md 的 MS-004/Criterion 2(引用,不複述;不是「最高層級」)
|
|
130
132
|
|
|
131
|
-
#### CONVERGE Phase(BQS 第 2 層 — D5–D8
|
|
132
|
-
1. 以 3
|
|
133
|
-
2.
|
|
134
|
-
3.
|
|
135
|
-
4. **Meta
|
|
133
|
+
#### CONVERGE Phase(BQS 第 2 層 — D5–D8 僅施於推薦集)
|
|
134
|
+
1. 以 3 個獨立評審(工程可行性/使用者影響/策略一致)各自評分後聚合;**評審提示以「目標+約束」表達**(判準,不是步驟清單——理由見 skill 引用的 core/model-selection.md MS-009),加權公式與逐準則指南為參考格式
|
|
135
|
+
2. 對**推薦集**(Agg. Score ≥ 3.5,不限筆數)跑硬角色 Devil's Advocate(論證會失敗)+ Steelman;使用者以 (a)修改/(b)反駁/(c)移除 回應。**D4 判官≠產生者,兩軸**:(1) 評審在獨立 context(`--enhanced`)執行,且 (2) 宣告模型層級(core/model-selection.md 的廠商中立標籤;指名具體廠商模型即違規)——缺任一軸標 `[degraded]` 不得 pass
|
|
136
|
+
3. **僅對推薦集**套用 D5 接地(外部事實需 file:line/來源、跨級地板;假說免接地)、D6 淨值(解誰問題/真有嗎/不做代價+掛 lagging 欄)、D7 二態證偽(「需先做 X」轉 next-step、不算 fail)、D8 next-step 裁決——四者為判準、不限順序
|
|
137
|
+
4. **Meta 停止規則**:維度全綠且再跑一輪推薦集集合成員不變→停;硬上限 2 輪
|
|
136
138
|
|
|
137
139
|
🛑 **STOP**: 展示 Brainstorm Report(含 Seeds/爭議區/Judgment Override)後等待使用者決定下一步
|
|
138
140
|
|
|
@@ -153,13 +155,15 @@ PRE-FLIGHT ──► FRAME ──► DIVERGE ──────────► C
|
|
|
153
155
|
| 使用「像 X 但給 Y」種子 | 改寫為底層問題再繼續(反種子 guardrail) |
|
|
154
156
|
| 指定的 persona/透鏡不存在 | 列出可用選項供選擇 |
|
|
155
157
|
| `--enhanced` 但宿主不支援子代理 | 靜默退回 baseline,照常進行 |
|
|
156
|
-
|
|
|
158
|
+
| 推薦集全來自同一 persona/透鏡 | 標示並在輸出前再跑一個透鏡 |
|
|
157
159
|
| baseline 單 context 跑評審(無獨立 context) | BQS D4 標 `[degraded]`,不得標 pass |
|
|
160
|
+
| 評審未宣告模型層級(或以具體廠商模型名宣告) | BQS D4 標 `[degraded]` 不得 pass;模型名改為 core/model-selection.md 的層級標籤 |
|
|
158
161
|
| 外部事實宣稱無 file:line/來源 | D5 接地 fail,要求補來源或改標 `[假說]` |
|
|
159
162
|
|
|
160
163
|
## 參考
|
|
161
164
|
|
|
162
165
|
* [Brainstorm Assistant Skill](../brainstorm-assistant/SKILL.md)
|
|
163
166
|
* [Brainstorm Assistant Guide](../brainstorm-assistant/guide.md)
|
|
167
|
+
* [Model Selection](../../core/model-selection.md) — BQS D4 與發散側的模型層級標籤(引用,不複述)
|
|
164
168
|
* [Requirement Assistant](../requirement-assistant/SKILL.md)
|
|
165
169
|
* [Spec-Driven Development](../spec-driven-dev/SKILL.md)
|
|
@@ -42,7 +42,7 @@ argument-hint: "[--level N | --yes]"
|
|
|
42
42
|
| **OpenCode** | `AGENTS.md` | ✅ | ✅ |
|
|
43
43
|
| **Gemini CLI** | `GEMINI.md` | ✅ | ✅ |
|
|
44
44
|
| **Codex** | `AGENTS.md` | ✅ | ❌ |
|
|
45
|
-
| **Antigravity** |
|
|
45
|
+
| **Antigravity** | `.agents/AGENTS.md` | ✅ | ❌ |
|
|
46
46
|
|
|
47
47
|
預選環境中偵測到的工具。注意:Codex 和 OpenCode 共用 `AGENTS.md`。
|
|
48
48
|
|
|
@@ -211,7 +211,6 @@ Options:
|
|
|
211
211
|
| 選項 | 說明 |
|
|
212
212
|
|------|------|
|
|
213
213
|
| **Standard(建議)** | 摘要 + 任務對應,AI 知道何時讀取哪項標準 |
|
|
214
|
-
| **Full Embed** | 嵌入所有規則,AI 可立即使用但檔案較大 |
|
|
215
214
|
| **Minimal** | 僅檔案參考,適合搭配 Skills 使用 |
|
|
216
215
|
|
|
217
216
|
### 步驟 13:確認並執行
|
|
@@ -126,6 +126,8 @@ Closes #123
|
|
|
126
126
|
|
|
127
127
|
## 參考
|
|
128
128
|
|
|
129
|
+
- [Conventional Commits 指南](./conventional-commits.md) — 需要確切的 type、scope 與格式規則時讀它。
|
|
130
|
+
- [Commit Message 語言選項](./language-options.md) — 要決定 commit message 用哪一種語言時讀它。
|
|
129
131
|
- 詳細指南:[guide.md](./guide.md)
|
|
130
132
|
- 核心規範:[commit-message-guide.md](../../../../core/commit-message-guide.md)
|
|
131
133
|
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
name: contract-test-assistant
|
|
3
3
|
source: ../../../../skills/contract-test-assistant/SKILL.md
|
|
4
4
|
source_version: 1.0.0
|
|
5
|
-
source_hash:
|
|
5
|
+
source_hash: 35b654f0b277
|
|
6
6
|
translation_version: 1.1.0
|
|
7
7
|
last_synced: 2026-08-17
|
|
8
8
|
status: current
|
|
@@ -103,6 +103,7 @@ ASSESS ──► CHOOSE ──► DEFINE ──► IMPLEMENT ──► VERIFY
|
|
|
103
103
|
|
|
104
104
|
## 參考
|
|
105
105
|
|
|
106
|
+
- 詳細指南:[guide.md](./guide.md) — 合約測試的策略與實作。要設定 Pact/OpenAPI,或把合約測試接進 CI 時讀它。
|
|
106
107
|
- 詳細指南:[contract-testing.md](../../options/testing/contract-testing.md)
|
|
107
108
|
- 相關:[api-design-assistant](../api-design-assistant/SKILL.md)
|
|
108
109
|
|
|
@@ -47,6 +47,7 @@ description: |
|
|
|
47
47
|
|
|
48
48
|
## 參考
|
|
49
49
|
|
|
50
|
+
- 詳細指南:[guide.md](./guide.md) — 日誌、指標、追蹤三大支柱的完整說明。要建立監控、挑選指標型別、導入追蹤或評估成熟度時讀它。
|
|
50
51
|
- 核心規範:[observability-standards.md](../../../../core/observability-standards.md)
|
|
51
52
|
- 核心規範:[alerting-standards.md](../../../../core/alerting-standards.md)
|
|
52
53
|
- 核心規範:[slo-standards.md](../../../../core/slo-standards.md)
|
|
@@ -151,6 +151,7 @@ project/
|
|
|
151
151
|
|
|
152
152
|
## 相關標準
|
|
153
153
|
|
|
154
|
+
- [各語言的專案結構](./language-patterns.md) — 逐語言的實際目錄配置(Node.js/TypeScript、Python、Go 等)。要為某個特定語言擺放專案結構時讀它。
|
|
154
155
|
- [核心:專案結構](../../core/project-structure.md)
|
|
155
156
|
- [核心:文件結構](../../core/documentation-structure.md)
|
|
156
157
|
- [指南:檔案歸檔決策指南](../../core/guides/file-placement-guide.md)
|
|
@@ -71,5 +71,8 @@ description: |
|
|
|
71
71
|
|
|
72
72
|
## 參考
|
|
73
73
|
|
|
74
|
+
- [變更日誌格式指南](./changelog-format.md) — 要撰寫或審查變更日誌時讀它。
|
|
75
|
+
- [發布流程指南](./release-workflow.md) — 真的要發一版、需要逐步流程時讀它。
|
|
76
|
+
- [語意化版本指南](./semantic-versioning.md) — 要判斷一個變更算 major、minor 還是 patch 時讀它。
|
|
74
77
|
- 詳細指南:[guide.md](./guide.md)
|
|
75
78
|
- 核心規範:[versioning.md](../../../../core/versioning.md)
|
|
@@ -74,5 +74,7 @@ So that [benefit].
|
|
|
74
74
|
|
|
75
75
|
## 參考
|
|
76
76
|
|
|
77
|
+
- [需求完整性檢查清單](./requirement-checklist.md) — 要在需求被接受之前檢查它完不完整時讀它。
|
|
78
|
+
- [需求撰寫指南](./requirement-writing.md) — 要從零開始寫一份需求時讀它。
|
|
77
79
|
- 詳細指南:[guide.md](./guide.md)
|
|
78
80
|
- 核心規範:[requirement-engineering.md](../../../../core/requirement-engineering.md)
|
|
@@ -126,6 +126,9 @@ description: |
|
|
|
126
126
|
|
|
127
127
|
## 參考
|
|
128
128
|
|
|
129
|
+
- [BDD 提取工作流程指南](./bdd-extraction.md) — 要從既有規格裡萃取 BDD 場景時讀它。
|
|
130
|
+
- [TDD 分析工作流程指南](./tdd-analysis.md) — 要對照場景分析測試涵蓋率、找出缺口時讀它。
|
|
131
|
+
- 逐步流程:[workflow.md](./workflow.md) — 逆向工程的各階段與順序(程式碼掃描、資料模型、設定、測試分析、涵蓋率)。真的要跑一次逆向時讀它。
|
|
129
132
|
- 詳細指南:[guide.md](./guide.md)
|
|
130
133
|
- 核心規範:[reverse-engineering-standards.md](../../../../core/reverse-engineering-standards.md)
|
|
131
134
|
|
|
@@ -605,9 +605,19 @@ test.each([
|
|
|
605
605
|
|
|
606
606
|
## CI/CD 整合
|
|
607
607
|
|
|
608
|
-
### GitHub Actions
|
|
608
|
+
### GitHub Actions —— 目前無法腳本化
|
|
609
|
+
|
|
610
|
+
`uds reverse-tdd` 這個 CLI 指令不存在。覆蓋率缺口分析是以 **`reverse-tdd`
|
|
611
|
+
AI agent**(`uds agent install reverse-tdd`)提供——由你的 AI 工具互動式執行,
|
|
612
|
+
不是會吐出 `coverage.json` 的確定性腳本。本文件先前版本示範了一個呼叫不存在
|
|
613
|
+
CLI 指令的 GitHub Actions 步驟(XSPEC-383 R4,2026-08-19)。
|
|
614
|
+
|
|
615
|
+
下方僅為**示意**——展示若未來真的做出可腳本化的分析器,CI 整合可能長什麼樣,
|
|
616
|
+
不是現在就能貼進 workflow 使用的東西:
|
|
609
617
|
|
|
610
618
|
```yaml
|
|
619
|
+
# 示意用途——目前沒有可腳本化的等價物。
|
|
620
|
+
# 現況:改用 reverse-tdd agent 從你的 AI 工具互動式執行。
|
|
611
621
|
name: BDD Coverage Check
|
|
612
622
|
|
|
613
623
|
on: [pull_request]
|
|
@@ -618,29 +628,9 @@ jobs:
|
|
|
618
628
|
steps:
|
|
619
629
|
- uses: actions/checkout@v4
|
|
620
630
|
|
|
621
|
-
- name: Run BDD Coverage Analysis
|
|
622
|
-
run: |
|
|
623
|
-
uds reverse-tdd features/ --output coverage.json
|
|
624
|
-
|
|
625
|
-
- name: Check Coverage Threshold
|
|
631
|
+
- name: Run BDD Coverage Analysis (尚未實作)
|
|
626
632
|
run: |
|
|
627
|
-
|
|
628
|
-
if (( $(echo "$COVERAGE < 0.80" | bc -l) )); then
|
|
629
|
-
echo "Coverage $COVERAGE below 80% threshold"
|
|
630
|
-
exit 1
|
|
631
|
-
fi
|
|
632
|
-
|
|
633
|
-
- name: Comment on PR
|
|
634
|
-
uses: actions/github-script@v6
|
|
635
|
-
with:
|
|
636
|
-
script: |
|
|
637
|
-
const coverage = require('./coverage.json');
|
|
638
|
-
github.rest.issues.createComment({
|
|
639
|
-
issue_number: context.issue.number,
|
|
640
|
-
owner: context.repo.owner,
|
|
641
|
-
repo: context.repo.repo,
|
|
642
|
-
body: `## BDD 覆蓋率報告\n\n覆蓋率: ${(coverage.summary.effectiveCoverage * 100).toFixed(1)}%`
|
|
643
|
-
});
|
|
633
|
+
echo "目前沒有 CLI 等價物——互動式用法請見 reverse-tdd agent"
|
|
644
634
|
```
|
|
645
635
|
|
|
646
636
|
---
|
|
@@ -46,6 +46,7 @@ description: |
|
|
|
46
46
|
|
|
47
47
|
## 參考
|
|
48
48
|
|
|
49
|
+
- 詳細指南:[guide.md](./guide.md) — 運維手冊的撰寫、組織與驗證。要寫手冊、審查運維程序、規劃演練或評估涵蓋範圍時讀它。
|
|
49
50
|
- 核心規範:[runbook-standards.md](../../../../core/runbook-standards.md)
|
|
50
51
|
- 相關:[alerting-standards.md](../../../../core/alerting-standards.md)
|
|
51
52
|
- 相關:[postmortem-standards.md](../../../../core/postmortem-standards.md)
|
|
@@ -46,6 +46,7 @@ description: |
|
|
|
46
46
|
|
|
47
47
|
## 參考
|
|
48
48
|
|
|
49
|
+
- 詳細指南:[guide.md](./guide.md) — SLI/SLO/錯誤預算的定義與管理。要定義 SLO、挑選 SLI、計算錯誤預算或設定可靠度目標時讀它。
|
|
49
50
|
- 核心規範:[slo-standards.md](../../../../core/slo-standards.md)
|
|
50
51
|
- 相關:[observability-standards.md](../../../../core/observability-standards.md)
|
|
51
52
|
- 相關:[alerting-standards.md](../../../../core/alerting-standards.md)
|
|
@@ -164,18 +164,9 @@ steps:
|
|
|
164
164
|
|
|
165
165
|
## 用法
|
|
166
166
|
|
|
167
|
-
###
|
|
167
|
+
### 取得 Workflows
|
|
168
168
|
|
|
169
|
-
|
|
170
|
-
# List available workflows
|
|
171
|
-
uds workflow list
|
|
172
|
-
|
|
173
|
-
# Install specific workflow
|
|
174
|
-
uds workflow install integrated-flow
|
|
175
|
-
|
|
176
|
-
# Install all workflows
|
|
177
|
-
uds workflow install --all
|
|
178
|
-
```
|
|
169
|
+
`uds workflow` 已於 v6.0.0 移除(流程編排職責移至採用層——DEC-049;見 [MIGRATION-v6.md §3](../../../../docs/MIGRATION-v6.md#3-removed-4-deprecated-cli-commands))。沒有 CLI 安裝步驟:workflow 定義直接存放在本 repo 的 `skills/workflows/*.workflow.yaml`(見上方表格),請直接引用路徑或複製到專案的 `.claude/workflows/`。
|
|
179
170
|
|
|
180
171
|
### 執行
|
|
181
172
|
|
|
@@ -97,6 +97,8 @@ After `/atdd` completes, the AI assistant should suggest:
|
|
|
97
97
|
|
|
98
98
|
## Reference | 參考
|
|
99
99
|
|
|
100
|
+
- [Acceptance Criteria Writing Guide](./acceptance-criteria-guide.md) — Read it when writing or reviewing acceptance criteria.
|
|
101
|
+
- [ATDD Workflow Guide](./atdd-workflow.md) — Read it when running the ATDD cycle end to end.
|
|
100
102
|
- Detailed guide: [guide.md](./guide.md)
|
|
101
103
|
- Core standard: [acceptance-test-driven-development.md](../../core/acceptance-test-driven-development.md)
|
|
102
104
|
|
|
@@ -92,6 +92,8 @@ After `/bdd` completes, the AI assistant should suggest:
|
|
|
92
92
|
|
|
93
93
|
## Reference | 參考
|
|
94
94
|
|
|
95
|
+
- [BDD Workflow Guide](./bdd-workflow.md) — Read it when running discovery, formulation and automation as a cycle.
|
|
96
|
+
- [Gherkin Quick Reference Guide](./gherkin-guide.md) — Read it when writing Given/When/Then and you need the exact syntax.
|
|
95
97
|
- Detailed guide: [guide.md](./guide.md)
|
|
96
98
|
- Core standard: [behavior-driven-development.md](../../core/behavior-driven-development.md)
|
|
97
99
|
|
|
@@ -29,6 +29,14 @@ v3 把發散從「單一 AI 衝數量」改為**persona 集成**(每個角色
|
|
|
29
29
|
|
|
30
30
|
**v4 的核心改動:** v3 提供了強健的*機制*,卻沒有可判定 pass/fail 的品質閘。v4 在其上疊加一道**時序化品質契約**——**腦力激盪品質標準(BQS v1)**——且不移除任何 v3 行為。BQS 是**四層 × 時間軸**結構:第 0 層宣告 explore/exploit 意圖;第 1 層(過程、leading、發散期全程可見)跑維度 **D1–D4**;第 2 層(產物、leading、收斂後僅施於推薦集,即 Agg. Score ≥ 3.5、不限筆數的想法)跑 **D5–D8** 加 Seeds 欄與爭議區;第 3 層是不受治理的 **Judgment Override**。第一原理:**決策用 leading 訊號、校準用 lagging 訊號——兩者是時間前後,而非對錯取捨。** 硬序列閘禁止 D5–D8 在發散期被揭示或評分。詳見下方「BQS v1 — 品質契約」。
|
|
31
31
|
|
|
32
|
+
**What changed in v4.2 (XSPEC-388) | v4.2 的核心改動:** D4 pinned where the judge sits and said nothing about whether it can think. An independent context running a literal-following model satisfies the old wording verbatim and rubber-stamps its way through D5–D8 — indistinguishable on paper from a review where someone actually objected. v4.2 gives D4 a **second axis**: the review must also declare its required **model level** (see the terminology note below). The CONVERGE briefs are re-shaped as **goal + constraints** — D5–D8 are criteria the judgment must satisfy, not steps to execute in order; the fixed sentence patterns and per-criterion scoring guides are kept as reference formats. The divergence side cites [core/model-selection.md](../../core/model-selection.md) (Standard or higher) instead of adding a rule of its own. No v3 mechanism is removed.
|
|
33
|
+
|
|
34
|
+
**v4.2 的核心改動(XSPEC-388):** D4 原本只釘住判官坐在哪裡,一個字都沒說它得想得動。一個獨立 context 跑字面遵循型模型會逐字滿足舊條文,然後照 D5–D8 逐條打勾放行——與一場真的有人反對過的評審在報告上無從分辨。v4.2 給 D4 加上**第二個軸**:評審還必須宣告其**模型層級**要求(見下方術語說明)。CONVERGE 的提示改為**目標+約束**形狀——D5–D8 是判斷必須滿足的判準,不是按順序執行的步驟;既有固定句型與逐準則分數指南保留為參考格式。發散側引用 [core/model-selection.md](../../core/model-selection.md)(Standard 或更高)而不另立規則。不移除任何 v3 機制。
|
|
35
|
+
|
|
36
|
+
> **Terminology — "model level", never "tier" | 術語——「模型層級」,不用「tier」:** inside this skill, `tier` is already taken twice — the BQS tiering table (which dimensions apply) and the Enhanced Tier. The model-capability axis defined in [core/model-selection.md](../../core/model-selection.md) (called *model tier* there) is therefore always written **model level(模型層級)** in this skill. Model levels are that standard's **vendor-neutral labels**; this skill never names a concrete vendor model (see D4).
|
|
37
|
+
>
|
|
38
|
+
> 在本 skill 內,`tier` 已被佔用兩次——BQS 分級表(套用哪些維度)與 Enhanced Tier。[core/model-selection.md](../../core/model-selection.md) 定義的模型能力軸(該標準稱 *model tier*)在本 skill 一律寫作**模型層級(model level)**。模型層級是該標準的**廠商中立標籤**;本 skill 絕不指名具體廠商模型(見 D4)。
|
|
39
|
+
|
|
32
40
|
## Mode Selection | 使用前先選模式
|
|
33
41
|
|
|
34
42
|
Apply these **objective triggers** before starting. Default is full v3 — routing rules are shortcuts to skip phases, not barriers to add.
|
|
@@ -81,7 +89,7 @@ At the start, declare the **explore/exploit ratio** and the bet type (incrementa
|
|
|
81
89
|
| **D1 Frame purity** | Problem embeds a specific solution or a "like X but for Y" framing → fail; must run 5-Whys to the root cause | 問題內嵌特定方案或「像 X 給 Y」→fail;須 5-Whys 到根因 |
|
|
82
90
|
| **D2 Divergence coverage** | Surviving ideas span < 3 independent personas/lenses → fail (modulated by Layer 0 weight); only a round with zero new ideas counts as saturation | 存活想法跨 <3 獨立 persona/透鏡→fail(受第 0 層權重調節);連一輪零新增才算飽和 |
|
|
83
91
|
| **D3 Cross-session diversity** | Seed is a competitor analogy → fail; Top all from a single lens → fail | 種子是競品類比→fail;Top 全來自單一 lens→fail |
|
|
84
|
-
| **D4 Evaluation de-bias** | Requires ≥3 critics + hard-role Devil's Advocate + Steelman; **pass requires an independent context
|
|
92
|
+
| **D4 Evaluation de-bias** | Requires ≥3 critics + hard-role Devil's Advocate + Steelman; **pass requires both axes**: (1) an independent context, **and** (2) a declared **model level** requirement using the vendor-neutral labels of [model-selection](../../core/model-selection.md) — naming a concrete vendor model is a violation. Missing either axis → mark `[degraded]` — must NOT be marked pass | 須 ≥3 評審 + 硬角色 Devil's Advocate + Steelman;**pass 須同時滿足兩軸**:(1) 獨立 context,**且** (2) 以 [model-selection](../../core/model-selection.md) 的廠商中立標籤宣告**模型層級**要求——指名具體廠商模型即違規。缺任一軸→標 `[degraded]` 不得標 pass |
|
|
85
93
|
|
|
86
94
|
> **Hard sequence gate | 硬序列閘:** D5–D8 are **forbidden** from being revealed or scored during the divergence period. The CONVERGE critics MUST NOT be invoked before the **last persona** has produced its set.
|
|
87
95
|
>
|
|
@@ -109,7 +117,7 @@ A reserved space no oracle enters: a human's intuition to "keep / kill" needs on
|
|
|
109
117
|
### Structural rules | 結構規則
|
|
110
118
|
|
|
111
119
|
1. **Meta stop rule | Meta 停止規則:** stop when the layer's dimensions are all green **and**, after one more round, the **Recommended Set membership is unchanged** (set membership, not internal ranking). **Hard cap: 2 rounds.** This replaces the vague "decision didn't flip" criterion. | 該層維度全綠 **且** 再跑一輪後 **推薦集成員不變**(看集合成員、不看內部排序)→停。**硬上限 2 輪**。取代模糊的「不翻決策」。
|
|
112
|
-
2. **Judge ≠ generator | 判官≠產生者:** D2/D4/D5/D7 judgements need an independent viewpoint; single-context self-evaluation may only be `[degraded]`, never pass. | D2/D4/D5/D7 判定須獨立視角;單 context 自評只能 `[degraded]`,不得 pass。
|
|
120
|
+
2. **Judge ≠ generator | 判官≠產生者:** D2/D4/D5/D7 judgements need an independent viewpoint; single-context self-evaluation may only be `[degraded]`, never pass. D4 additionally requires a declared model level — independence is one of its two axes (see D4). | D2/D4/D5/D7 判定須獨立視角;單 context 自評只能 `[degraded]`,不得 pass。D4 另需宣告模型層級——獨立性只是它兩軸之一(見 D4)。
|
|
113
121
|
3. **Calibration loop (consumes lagging) | 校準回路(吃 lagging):** v3's three Session Self-Evaluation metrics are folded in as the **lagging** end — **Adoption Rate = D6 lagging validation, Diversity = D2/D3 lagging observation, Cognitive Load = a cost constraint**. **Two parallel evaluation systems are forbidden** — there is one loop, not a separate self-eval. | v3 的 **Adoption Rate**=D6 滯後驗證、**Diversity**=D2/D3 滯後觀測、**Cognitive Load**=成本約束;三者收編為此回路 lagging 端,**禁兩套平行評估**。
|
|
114
122
|
4. **BQS self-evolution (lightweight) | BQS 自我演化(輕量):** versioned + an evidence `last-reviewed` that flags when overdue; "periodically brainstorm BQS itself" is optional. | 版本化 + 證據清單 last-reviewed 逾期 flag;「定期對 BQS 自身 brainstorm」列可選。
|
|
115
123
|
5. **Minimal sufficiency | 最小充分原則:** apply only the fewest dimensions the tier requires + confining Layer 2 to the Recommended Set (Agg. Score ≥ 3.5) instead of every divergence idea — this is the gate for cognitive economy; no separate cost dimension is added. | 套用該 tier 所需的最少維度 + 把第 2 層限縮在推薦集(Agg. Score ≥ 3.5),而非全部發散想法——認知經濟性的守門,不另設成本維度。
|
|
@@ -180,6 +188,10 @@ Define the problem space clearly before generating ideas.
|
|
|
180
188
|
>
|
|
181
189
|
> **v3 核心機制:** **persona 集成**——每個 persona 以**思維鏈**在**隔離**狀態下推理——再乘上**多樣性透鏡**。Meincke、Mollick、Terwiesch(2024)發現「思維鏈 + persona」的想法多樣性高於所有受測提示策略,接近人類團體。光衝數量是弱代理;結構性逼出不同視角才是真正槓桿。
|
|
182
190
|
|
|
191
|
+
> **Model level (divergence side) | 模型層級(發散側):** a divergence brief gives only a goal and constraints — exactly the trigger of [MS-004](../../core/model-selection.md#rules) / [Criterion 2](../../core/model-selection.md#criterion-2--specification-definiteness規格明確度) in core/model-selection.md: run divergence at **Standard or higher**. Not "the highest level" — that would overstate both what MS-004 says and our evidence. Lagging registry: this choice is validated through the calibration loop's existing **Diversity (D2/D3 lagging observation)** column — no second registry.
|
|
192
|
+
>
|
|
193
|
+
> 發散提示只給目標與約束——正是 core/model-selection.md 的 [MS-004](../../core/model-selection.md#rules)/[Criterion 2](../../core/model-selection.md#criterion-2--specification-definiteness規格明確度) 的觸發條件:發散以 **Standard 或更高**層級執行。不是「最高層級」——那會同時超出 MS-004 所說的與我們的證據。lagging 登記:此選擇由校準回路既有的 **Diversity(D2/D3 滯後觀測)** 欄驗證,不另開第二套登記。
|
|
194
|
+
|
|
183
195
|
#### Step 2a — Persona ensemble | persona 集成
|
|
184
196
|
|
|
185
197
|
Generate ideas through a default ensemble of personas. Each persona reasons **step by step (chain-of-thought)** and produces 2–4 ideas **from its own lens only**. The user may add, drop, or rename personas via `--personas`.
|
|
@@ -238,11 +250,15 @@ The "best ideas appear in the second half" pattern (Nijstad) is a **human-group*
|
|
|
238
250
|
>
|
|
239
251
|
> **v3 核心機制:** **多評審面板**取代單一加權評分者。單一 LLM 是弱且有偏的評估者(Li 等,2025:LLM 強於生成/精煉、弱於評估——人類保留最終裁決權)。三個獨立評審透鏡各自評分後聚合。
|
|
240
252
|
|
|
253
|
+
> **Critic brief shape — goal + constraints (v4.2) | 評審提示形狀——目標+約束(v4.2):** when dispatching the panel, the Devil's Advocate, or the Steelman — above all into independent contexts (`--enhanced`) — the brief MUST be phrased as a **goal plus the constraints the judgment must satisfy**, never as an ordered procedure. D5–D8 and the specificity requirement below are **criteria** ("your judgment must satisfy these"), not **steps** ("do these in this order"). The fixed sentence patterns, the per-criterion 1–5 guides, and the critic weight formulas below stay available as **reference formats** — none of them is a mandatory step. Rationale: [MS-009](../../core/model-selection.md#rules) in core/model-selection.md.
|
|
254
|
+
>
|
|
255
|
+
> 對面板、Devil's Advocate、Steelman 派工時——尤其派入獨立 context(`--enhanced`)——提示**必須**以**目標+其判斷必須滿足的約束**表達,不得寫成按順序執行的步驟清單。D5–D8 與下方的具體性要求是**判準**(「你的判斷必須滿足這些」),不是**步驟**(「照這個順序做這些事」)。既有固定句型、逐準則 1–5 指南與評審權重公式**保留為參考格式**,皆非必經步驟。理由:core/model-selection.md 的 [MS-009](../../core/model-selection.md#rules)。
|
|
256
|
+
|
|
241
257
|
#### Step 3a: Multi-critic panel | 多評審面板
|
|
242
258
|
|
|
243
|
-
Run **three independent critics**, each scoring every idea 1–5 on its own lens. Aggregate (mean) to reduce single-critic bias.
|
|
259
|
+
Run **three independent critics**, each scoring every idea 1–5 on its own lens. Aggregate (mean) to reduce single-critic bias. The weighted criteria below define the default lenses — a reference format, not a script the critic must walk through.
|
|
244
260
|
|
|
245
|
-
跑**三個獨立評審**,各自以自己的透鏡對每個想法打 1–5
|
|
261
|
+
跑**三個獨立評審**,各自以自己的透鏡對每個想法打 1–5 分;取平均聚合以降低單評審偏誤。下方加權準則定義預設透鏡——是參考格式,不是評審必須逐步照走的腳本。
|
|
246
262
|
|
|
247
263
|
| Critic lens | Weighted criteria it owns | 評審透鏡 |
|
|
248
264
|
|-------------|---------------------------|----------|
|
|
@@ -250,7 +266,7 @@ Run **three independent critics**, each scoring every idea 1–5 on its own lens
|
|
|
250
266
|
| **User impact** | Impact 70% · Alignment 30% | 使用者影響 |
|
|
251
267
|
| **Strategic alignment** | Alignment 60% · Impact 40% | 策略一致性 |
|
|
252
268
|
|
|
253
|
-
Per-criterion guide (1–5): Feasibility (5=trivial … 1=near-impossible); Impact (5=transformative … 1=negligible); Effort (5=hours … 1=quarters, inverted so lower effort scores higher); Alignment (5=core mission … 1=off-mission).
|
|
269
|
+
Per-criterion guide (1–5, reference format): Feasibility (5=trivial … 1=near-impossible); Impact (5=transformative … 1=negligible); Effort (5=hours … 1=quarters, inverted so lower effort scores higher); Alignment (5=core mission … 1=off-mission).
|
|
254
270
|
|
|
255
271
|
> **Optional — RICE / ICE (product features):** for prioritising shippable features, score `RICE = (Reach × Impact × Confidence) / Effort` or the lighter `ICE = Impact × Confidence × Ease`. Let engineers — not the LLM — estimate Effort; the LLM lacks codebase knowledge. RICE favours incremental wins, so don't use it alone for strategic bets.
|
|
256
272
|
>
|
|
@@ -262,9 +278,9 @@ A soft "please critique this" instruction yields mostly agreement (sycophancy).
|
|
|
262
278
|
|
|
263
279
|
軟性「請批評一下」只會得到附和(諂媚)。v3 指派**硬角色**:對**推薦集**(Agg. Score ≥ 3.5,不限筆數)中的每一個想法各跑一個 **Devil's Advocate**(「你的任務是論證此案會失敗」)與一個 **Steelman**(「說出反方最強而善意的版本」)。兩者一起壓力測試韌性,而非只是戳。
|
|
264
280
|
|
|
265
|
-
|
|
281
|
+
Constraint: each counterargument must name a **specific context** and a **specific reason** — vague concerns ("this might be hard") are rejected. The sentence pattern "This idea will fail in [specific context] because [specific reason]" is a reference format, not a required phrasing.
|
|
266
282
|
|
|
267
|
-
|
|
283
|
+
約束:每個反對理由須指出**具體情境**與**具體原因**——模糊顧慮(「這可能有點難」)不接受。句型「在 [具體情境] 下,此想法會失敗,因為 [具體原因]」是參考格式,不是必用句式。
|
|
268
284
|
|
|
269
285
|
The user **must** respond to each before advancing:
|
|
270
286
|
|
|
@@ -278,20 +294,21 @@ The user **must** respond to each before advancing:
|
|
|
278
294
|
|
|
279
295
|
**Flag:** `--no-rebuttal` skips this step; report section marked "Rebuttal: skipped".
|
|
280
296
|
|
|
281
|
-
> **BQS D4 — judge ≠ generator | BQS D4 —
|
|
297
|
+
> **BQS D4 — judge ≠ generator, two axes | BQS D4 — 判官≠產生者,兩軸:** D4 passes **only when both axes hold**. **(1) Seat — independence**: the critics/Devil's Advocate run in an independent context (the `--enhanced` isolated-agent host). **(2) Mind — declared model level**: the review declares its required **model level**, expressed with the vendor-neutral labels of [core/model-selection.md](../../core/model-selection.md); selection basis is that standard's criteria (link, don't restate). Naming a concrete vendor model is a violation — the level labels are the only legal form. Missing either axis → the panel is marked `[degraded]` and **must NOT be marked pass**: an independent context running a literal-following model satisfies the seat axis verbatim and is a fully compliant rubber stamp — recognizing "the same thing under a different name" is judgment work that box-ticking cannot do. On a baseline single context, the panel is same-context self-evaluation — `[degraded]`, never pass, honest but not independent. Do not silently treat a baseline run as a D4 pass.
|
|
282
298
|
>
|
|
283
|
-
> D4
|
|
299
|
+
> D4 唯有**兩軸同時成立**才可 pass。**(1) 座位——獨立性**:評審/Devil's Advocate 在獨立 context 執行(`--enhanced` 隔離 agent 宿主)。**(2) 心智——宣告模型層級**:評審宣告其**模型層級**要求,以 [core/model-selection.md](../../core/model-selection.md) 的廠商中立標籤表達;選層依據為該標準的判準(引用,不複述)。指名具體廠商模型即違規——層級標籤是唯一合法的表達方式。缺任一軸→面板標 `[degraded]`、**不得標 pass**:獨立 context 跑字面遵循型模型會逐字滿足座位軸,成為一枚完全合規的橡皮圖章——認出「不同名字下的同一件事」是打勾做不到的判斷工作。baseline 單 context 下,面板是同 context 自評,標 `[degraded]`、不得 pass——誠實但非獨立。不得把 baseline 跑當成 D4 pass 靜默通過。
|
|
284
300
|
|
|
285
301
|
#### Step 3c: BQS Layer 2 — product gate on the Recommended Set | BQS 第 2 層——對推薦集的產物閘
|
|
286
302
|
|
|
287
|
-
After convergence, apply **D5–D8 to the Recommended Set only** (ideas with Agg. Score ≥ 3.5, uncapped — never to every divergence idea). If no idea reaches 3.5, keep the single highest-scoring idea and mark it `[below threshold — shown for reference]` so the report is never empty. For each idea in the Recommended Set:
|
|
303
|
+
After convergence, apply **D5–D8 to the Recommended Set only** (ideas with Agg. Score ≥ 3.5, uncapped — never to every divergence idea). If no idea reaches 3.5, keep the single highest-scoring idea and mark it `[below threshold — shown for reference]` so the report is never empty. D5–D8 are **criteria the judgment must satisfy** — in any order; they are not a procedure. For each idea in the Recommended Set, the judgment must satisfy:
|
|
288
304
|
|
|
289
|
-
收斂後,**僅對推薦集**(Agg. Score ≥ 3.5 的想法,不限筆數)套用 **D5–D8**(絕不對全部發散想法)。若無想法達 3.5,仍保留分數最高的 1 個,標記 `[未達門檻——僅供參考]
|
|
305
|
+
收斂後,**僅對推薦集**(Agg. Score ≥ 3.5 的想法,不限筆數)套用 **D5–D8**(絕不對全部發散想法)。若無想法達 3.5,仍保留分數最高的 1 個,標記 `[未達門檻——僅供參考]`,避免報告空白。D5–D8 是**判斷必須滿足的判準**——不限順序,不是程序。對推薦集中每個想法,判斷必須滿足:
|
|
290
306
|
|
|
291
307
|
- **D5 Grounding:** split each claim into `[current state / external fact]` (needs a file:line or source, else fail) vs `[future / hypothesis]` (no grounding needed, mark `[hypothesis]`). An **external-fact claim is a cross-tier floor** — it must be grounded even at the creative tier. | 將每個主張分流為 `[現狀/外部事實]`(需 file:line 或來源,否則 fail)vs `[未來/假說]`(免接地、標 `[假說]`)。**外部事實宣稱為跨級地板**——creative 級也須接地。
|
|
292
308
|
- **D6 Net benefit:** each idea in the Recommended Set must answer "whose problem / do we actually have it / cost of not doing it"; at least one idea must be eliminated as "not worth doing"; attach a **lagging registry field** (which later signal validates this). | 推薦集中每個想法須答「解誰問題/我們真有嗎/不做的代價」;至少一個淘汰為「不值得做」;掛 **lagging 登記欄**(事後哪個訊號驗證)。
|
|
293
309
|
- **D7 Falsifiability (two-state):** mark `[falsifiable now]` **or** `[need to do X first to define falsification]`. The latter is **routed to a next-step that feeds D8 — it does NOT count as fail**. | 標 `[現可陳述證偽]` **或** `[需先做 X 才能定義證偽]`。後者**轉 next-step 餵 D8——不算 fail**。
|
|
294
310
|
- **D8 Actionability:** every surviving idea needs a next-step decision (including an explicit "defer / not now"); none → fail. | 每個存活想法需 next-step 裁決(含明確「暫不做」);無 → fail。
|
|
311
|
+
- A `defer` next-step, and every idea parked in **Seeds (from killed ideas)**, are deferred items under [deferred-item-exit](../../core/deferred-item-exit.md). | `defer` 的 next-step,以及**Seeds(來自被淘汰的想法)**中每個被擱置的想法,皆為 [deferred-item-exit](../../core/deferred-item-exit.md) 所指的延後項目。
|
|
295
312
|
|
|
296
313
|
> **Meta stop rule (BQS structural rule 1) | Meta 停止規則:** stop iterating when the applied dimensions are all green **and** one more round leaves the **Recommended Set membership unchanged** (set membership, not internal ordering). **Hard cap: 2 rounds.**
|
|
297
314
|
>
|
|
@@ -350,9 +367,9 @@ Produce a Brainstorm Report ready for `/requirement` or `/sdd`. Each surviving i
|
|
|
350
367
|
- [ ] Proceed to `/sdd` if requirements are clear
|
|
351
368
|
```
|
|
352
369
|
|
|
353
|
-
> **BQS output additions (v4):** the **Seeds** section (rule: non-empty when ≥1 idea was killed), the **Contested Zone** (high-variance ideas not eliminated by mean ranking), and the **Judgment Override** channel (human decision overriding the aggregate score) are required by BQS Layer 2/3. The Recommendations block records the D5–D8 status per idea, for every idea in the Recommended Set (uncapped).
|
|
370
|
+
> **BQS output additions (v4):** the **Seeds** section (rule: non-empty when ≥1 idea was killed), the **Contested Zone** (high-variance ideas not eliminated by mean ranking), and the **Judgment Override** channel (human decision overriding the aggregate score) are required by BQS Layer 2/3. The Recommendations block records the D5–D8 status per idea, for every idea in the Recommended Set (uncapped). Since v4.2, the report header also records the review's declared **model level** (vendor-neutral label) — without it, D4 is `[degraded]` and the declaration has nowhere to be verified.
|
|
354
371
|
>
|
|
355
|
-
> **BQS 輸出新增(v4):** **Seeds** 區(規則:被殺 ≥1 則非空)、**爭議區**(高方差想法不按 mean 淘汰)、**Judgment Override** 通道(人類裁決凌駕聚合分)為 BQS 第 2/3 層所要求。Recommendations 區塊為推薦集(不限筆數)中每個想法逐項記錄 D5–D8
|
|
372
|
+
> **BQS 輸出新增(v4):** **Seeds** 區(規則:被殺 ≥1 則非空)、**爭議區**(高方差想法不按 mean 淘汰)、**Judgment Override** 通道(人類裁決凌駕聚合分)為 BQS 第 2/3 層所要求。Recommendations 區塊為推薦集(不限筆數)中每個想法逐項記錄 D5–D8 狀態。自 v4.2 起,報告表頭並記錄評審宣告的**模型層級**(廠商中立標籤)——沒有它,D4 為 `[degraded]`,且宣告無處可驗。
|
|
356
373
|
|
|
357
374
|
## Diversity-Collapse Guardrail | 多樣性崩塌防護
|
|
358
375
|
|
|
@@ -441,6 +458,7 @@ After `/brainstorm` completes, the AI assistant should suggest:
|
|
|
441
458
|
## Reference | 參考
|
|
442
459
|
|
|
443
460
|
- Detailed guide: [guide.md](./guide.md)
|
|
461
|
+
- Model level basis: [core/model-selection.md](../../core/model-selection.md) — the D4 mind axis and the divergence-side level defer to that standard's criteria. **Cite it, don't restate it** — a summary rots when its source is corrected; a pointer does not. | 模型層級依據:D4 心智軸與發散側層級皆以該標準的判準為準。**引用、不複述**——摘要會隨來源被更正而腐壞,指標不會。
|
|
444
462
|
|
|
445
463
|
## AI Agent Behavior | AI 代理行為
|
|
446
464
|
|