universal-dev-standards 6.6.0 → 6.7.0

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.
Files changed (177) hide show
  1. package/bundled/ai/standards/ai-response-navigation.ai.yaml +30 -3
  2. package/bundled/core/ai-response-navigation.md +62 -9
  3. package/bundled/locales/zh-CN/CHANGELOG.md +21 -2
  4. package/bundled/locales/zh-CN/README.md +1 -1
  5. package/bundled/locales/zh-CN/SECURITY.md +1 -1
  6. package/bundled/locales/zh-CN/core/ai-response-navigation.md +54 -8
  7. package/bundled/locales/zh-CN/skills/ac-coverage/SKILL.md +8 -4
  8. package/bundled/locales/zh-CN/skills/adr-assistant/SKILL.md +7 -6
  9. package/bundled/locales/zh-CN/skills/ai-collaboration-standards/SKILL.md +7 -2
  10. package/bundled/locales/zh-CN/skills/ai-friendly-architecture/SKILL.md +6 -5
  11. package/bundled/locales/zh-CN/skills/ai-instruction-standards/SKILL.md +6 -5
  12. package/bundled/locales/zh-CN/skills/api-design-assistant/SKILL.md +6 -5
  13. package/bundled/locales/zh-CN/skills/atdd-assistant/SKILL.md +6 -5
  14. package/bundled/locales/zh-CN/skills/audit-assistant/SKILL.md +6 -5
  15. package/bundled/locales/zh-CN/skills/bdd-assistant/SKILL.md +6 -5
  16. package/bundled/locales/zh-CN/skills/brainstorm-assistant/SKILL.md +7 -6
  17. package/bundled/locales/zh-CN/skills/changelog-guide/SKILL.md +6 -5
  18. package/bundled/locales/zh-CN/skills/checkin-assistant/SKILL.md +6 -5
  19. package/bundled/locales/zh-CN/skills/ci-cd-assistant/SKILL.md +6 -5
  20. package/bundled/locales/zh-CN/skills/code-review-assistant/SKILL.md +6 -5
  21. package/bundled/locales/zh-CN/skills/commit-standards/SKILL.md +7 -6
  22. package/bundled/locales/zh-CN/skills/contract-test-assistant/SKILL.md +7 -6
  23. package/bundled/locales/zh-CN/skills/database-assistant/SKILL.md +6 -5
  24. package/bundled/locales/zh-CN/skills/deploy-assistant/SKILL.md +7 -6
  25. package/bundled/locales/zh-CN/skills/dev-methodology/SKILL.md +7 -6
  26. package/bundled/locales/zh-CN/skills/dev-workflow-guide/SKILL.md +7 -6
  27. package/bundled/locales/zh-CN/skills/docs-generator/SKILL.md +6 -5
  28. package/bundled/locales/zh-CN/skills/documentation-guide/SKILL.md +6 -5
  29. package/bundled/locales/zh-CN/skills/durable-execution-assistant/SKILL.md +6 -5
  30. package/bundled/locales/zh-CN/skills/e2e-assistant/SKILL.md +7 -3
  31. package/bundled/locales/zh-CN/skills/error-code-guide/SKILL.md +5 -4
  32. package/bundled/locales/zh-CN/skills/git-workflow-guide/SKILL.md +6 -5
  33. package/bundled/locales/zh-CN/skills/incident-response-assistant/SKILL.md +6 -5
  34. package/bundled/locales/zh-CN/skills/journey-test-assistant/SKILL.md +7 -4
  35. package/bundled/locales/zh-CN/skills/knowledge-graph/SKILL.md +7 -4
  36. package/bundled/locales/zh-CN/skills/logging-guide/SKILL.md +7 -6
  37. package/bundled/locales/zh-CN/skills/metrics-dashboard-assistant/SKILL.md +6 -5
  38. package/bundled/locales/zh-CN/skills/migration-assistant/SKILL.md +7 -6
  39. package/bundled/locales/zh-CN/skills/observability-assistant/SKILL.md +7 -3
  40. package/bundled/locales/zh-CN/skills/orchestrate/SKILL.md +7 -6
  41. package/bundled/locales/zh-CN/skills/plan/SKILL.md +7 -6
  42. package/bundled/locales/zh-CN/skills/pr-automation-assistant/SKILL.md +6 -5
  43. package/bundled/locales/zh-CN/skills/project-discovery/SKILL.md +6 -5
  44. package/bundled/locales/zh-CN/skills/project-structure-guide/SKILL.md +7 -2
  45. package/bundled/locales/zh-CN/skills/push/SKILL.md +7 -6
  46. package/bundled/locales/zh-CN/skills/refactoring-assistant/SKILL.md +6 -5
  47. package/bundled/locales/zh-CN/skills/release-standards/SKILL.md +6 -5
  48. package/bundled/locales/zh-CN/skills/requirement-assistant/SKILL.md +6 -5
  49. package/bundled/locales/zh-CN/skills/retrospective-assistant/SKILL.md +6 -5
  50. package/bundled/locales/zh-CN/skills/reverse-engineer/SKILL.md +7 -6
  51. package/bundled/locales/zh-CN/skills/runbook-assistant/SKILL.md +7 -3
  52. package/bundled/locales/zh-CN/skills/security-assistant/SKILL.md +6 -5
  53. package/bundled/locales/zh-CN/skills/security-scan-assistant/SKILL.md +6 -5
  54. package/bundled/locales/zh-CN/skills/skill-builder/SKILL.md +7 -4
  55. package/bundled/locales/zh-CN/skills/slo-assistant/SKILL.md +7 -3
  56. package/bundled/locales/zh-CN/skills/spec-derivation/SKILL.md +7 -4
  57. package/bundled/locales/zh-CN/skills/spec-driven-dev/SKILL.md +7 -6
  58. package/bundled/locales/zh-CN/skills/sweep/SKILL.md +7 -6
  59. package/bundled/locales/zh-CN/skills/tdd-assistant/SKILL.md +6 -5
  60. package/bundled/locales/zh-CN/skills/test-coverage-assistant/SKILL.md +6 -5
  61. package/bundled/locales/zh-CN/skills/testing-guide/SKILL.md +7 -6
  62. package/bundled/locales/zh-TW/CHANGELOG.md +21 -2
  63. package/bundled/locales/zh-TW/README.md +1 -1
  64. package/bundled/locales/zh-TW/SECURITY.md +1 -1
  65. package/bundled/locales/zh-TW/core/ai-response-navigation.md +54 -8
  66. package/bundled/locales/zh-TW/skills/ac-coverage/SKILL.md +7 -3
  67. package/bundled/locales/zh-TW/skills/adr-assistant/SKILL.md +7 -3
  68. package/bundled/locales/zh-TW/skills/ai-collaboration-standards/SKILL.md +6 -5
  69. package/bundled/locales/zh-TW/skills/ai-friendly-architecture/SKILL.md +7 -3
  70. package/bundled/locales/zh-TW/skills/ai-instruction-standards/SKILL.md +7 -3
  71. package/bundled/locales/zh-TW/skills/api-design-assistant/SKILL.md +7 -3
  72. package/bundled/locales/zh-TW/skills/atdd-assistant/SKILL.md +7 -3
  73. package/bundled/locales/zh-TW/skills/audit-assistant/SKILL.md +7 -3
  74. package/bundled/locales/zh-TW/skills/bdd-assistant/SKILL.md +7 -3
  75. package/bundled/locales/zh-TW/skills/brainstorm-assistant/SKILL.md +8 -4
  76. package/bundled/locales/zh-TW/skills/changelog-guide/SKILL.md +7 -3
  77. package/bundled/locales/zh-TW/skills/checkin-assistant/SKILL.md +7 -3
  78. package/bundled/locales/zh-TW/skills/ci-cd-assistant/SKILL.md +7 -3
  79. package/bundled/locales/zh-TW/skills/code-review-assistant/SKILL.md +7 -3
  80. package/bundled/locales/zh-TW/skills/commit-standards/SKILL.md +7 -3
  81. package/bundled/locales/zh-TW/skills/contract-test-assistant/SKILL.md +7 -3
  82. package/bundled/locales/zh-TW/skills/database-assistant/SKILL.md +7 -3
  83. package/bundled/locales/zh-TW/skills/deploy-assistant/SKILL.md +6 -5
  84. package/bundled/locales/zh-TW/skills/dev-methodology/SKILL.md +6 -5
  85. package/bundled/locales/zh-TW/skills/dev-workflow-guide/SKILL.md +7 -3
  86. package/bundled/locales/zh-TW/skills/docs-generator/SKILL.md +7 -3
  87. package/bundled/locales/zh-TW/skills/documentation-guide/SKILL.md +7 -3
  88. package/bundled/locales/zh-TW/skills/durable-execution-assistant/SKILL.md +7 -3
  89. package/bundled/locales/zh-TW/skills/e2e-assistant/SKILL.md +7 -3
  90. package/bundled/locales/zh-TW/skills/error-code-guide/SKILL.md +7 -3
  91. package/bundled/locales/zh-TW/skills/git-workflow-guide/SKILL.md +7 -3
  92. package/bundled/locales/zh-TW/skills/incident-response-assistant/SKILL.md +7 -3
  93. package/bundled/locales/zh-TW/skills/journey-test-assistant/SKILL.md +7 -3
  94. package/bundled/locales/zh-TW/skills/knowledge-graph/SKILL.md +7 -3
  95. package/bundled/locales/zh-TW/skills/logging-guide/SKILL.md +8 -4
  96. package/bundled/locales/zh-TW/skills/metrics-dashboard-assistant/SKILL.md +7 -3
  97. package/bundled/locales/zh-TW/skills/migration-assistant/SKILL.md +7 -3
  98. package/bundled/locales/zh-TW/skills/observability-assistant/SKILL.md +7 -3
  99. package/bundled/locales/zh-TW/skills/orchestrate/SKILL.md +6 -5
  100. package/bundled/locales/zh-TW/skills/plan/SKILL.md +6 -5
  101. package/bundled/locales/zh-TW/skills/pr-automation-assistant/SKILL.md +7 -3
  102. package/bundled/locales/zh-TW/skills/project-discovery/SKILL.md +7 -3
  103. package/bundled/locales/zh-TW/skills/project-structure-guide/SKILL.md +6 -5
  104. package/bundled/locales/zh-TW/skills/push/SKILL.md +6 -5
  105. package/bundled/locales/zh-TW/skills/refactoring-assistant/SKILL.md +7 -3
  106. package/bundled/locales/zh-TW/skills/release-standards/SKILL.md +7 -3
  107. package/bundled/locales/zh-TW/skills/requirement-assistant/SKILL.md +7 -3
  108. package/bundled/locales/zh-TW/skills/retrospective-assistant/SKILL.md +7 -3
  109. package/bundled/locales/zh-TW/skills/reverse-engineer/SKILL.md +7 -3
  110. package/bundled/locales/zh-TW/skills/runbook-assistant/SKILL.md +7 -3
  111. package/bundled/locales/zh-TW/skills/security-assistant/SKILL.md +7 -3
  112. package/bundled/locales/zh-TW/skills/security-scan-assistant/SKILL.md +7 -3
  113. package/bundled/locales/zh-TW/skills/skill-builder/SKILL.md +7 -3
  114. package/bundled/locales/zh-TW/skills/slo-assistant/SKILL.md +7 -3
  115. package/bundled/locales/zh-TW/skills/spec-derivation/SKILL.md +7 -3
  116. package/bundled/locales/zh-TW/skills/spec-driven-dev/SKILL.md +7 -3
  117. package/bundled/locales/zh-TW/skills/sweep/SKILL.md +6 -5
  118. package/bundled/locales/zh-TW/skills/tdd-assistant/SKILL.md +7 -3
  119. package/bundled/locales/zh-TW/skills/test-coverage-assistant/SKILL.md +7 -3
  120. package/bundled/locales/zh-TW/skills/testing-guide/SKILL.md +8 -3
  121. package/bundled/skills/ac-coverage/SKILL.md +5 -1
  122. package/bundled/skills/adr-assistant/SKILL.md +1 -0
  123. package/bundled/skills/ai-collaboration-standards/SKILL.md +1 -0
  124. package/bundled/skills/ai-friendly-architecture/SKILL.md +1 -0
  125. package/bundled/skills/ai-instruction-standards/SKILL.md +1 -0
  126. package/bundled/skills/api-design-assistant/SKILL.md +1 -0
  127. package/bundled/skills/atdd-assistant/SKILL.md +5 -1
  128. package/bundled/skills/audit-assistant/SKILL.md +5 -1
  129. package/bundled/skills/bdd-assistant/SKILL.md +5 -1
  130. package/bundled/skills/brainstorm-assistant/SKILL.md +5 -1
  131. package/bundled/skills/changelog-guide/SKILL.md +5 -1
  132. package/bundled/skills/checkin-assistant/SKILL.md +5 -1
  133. package/bundled/skills/ci-cd-assistant/SKILL.md +1 -0
  134. package/bundled/skills/code-review-assistant/SKILL.md +5 -1
  135. package/bundled/skills/commit-standards/SKILL.md +5 -1
  136. package/bundled/skills/contract-test-assistant/SKILL.md +1 -0
  137. package/bundled/skills/database-assistant/SKILL.md +1 -0
  138. package/bundled/skills/deploy-assistant/SKILL.md +1 -0
  139. package/bundled/skills/dev-methodology/SKILL.md +5 -1
  140. package/bundled/skills/dev-workflow-guide/SKILL.md +5 -1
  141. package/bundled/skills/docs-generator/SKILL.md +5 -1
  142. package/bundled/skills/documentation-guide/SKILL.md +1 -0
  143. package/bundled/skills/durable-execution-assistant/SKILL.md +5 -1
  144. package/bundled/skills/e2e-assistant/SKILL.md +5 -1
  145. package/bundled/skills/error-code-guide/SKILL.md +1 -0
  146. package/bundled/skills/git-workflow-guide/SKILL.md +1 -0
  147. package/bundled/skills/incident-response-assistant/SKILL.md +1 -0
  148. package/bundled/skills/journey-test-assistant/SKILL.md +5 -1
  149. package/bundled/skills/knowledge-graph/SKILL.md +5 -1
  150. package/bundled/skills/logging-guide/SKILL.md +1 -0
  151. package/bundled/skills/metrics-dashboard-assistant/SKILL.md +5 -1
  152. package/bundled/skills/migration-assistant/SKILL.md +5 -1
  153. package/bundled/skills/observability-assistant/SKILL.md +1 -0
  154. package/bundled/skills/orchestrate/SKILL.md +1 -0
  155. package/bundled/skills/plan/SKILL.md +1 -0
  156. package/bundled/skills/pr-automation-assistant/SKILL.md +1 -0
  157. package/bundled/skills/project-discovery/SKILL.md +5 -1
  158. package/bundled/skills/project-structure-guide/SKILL.md +1 -0
  159. package/bundled/skills/push/SKILL.md +1 -0
  160. package/bundled/skills/refactoring-assistant/SKILL.md +5 -1
  161. package/bundled/skills/release-standards/SKILL.md +5 -1
  162. package/bundled/skills/requirement-assistant/SKILL.md +5 -1
  163. package/bundled/skills/retrospective-assistant/SKILL.md +1 -0
  164. package/bundled/skills/reverse-engineer/SKILL.md +5 -1
  165. package/bundled/skills/runbook-assistant/SKILL.md +1 -0
  166. package/bundled/skills/security-assistant/SKILL.md +1 -0
  167. package/bundled/skills/security-scan-assistant/SKILL.md +1 -0
  168. package/bundled/skills/skill-builder/SKILL.md +5 -1
  169. package/bundled/skills/slo-assistant/SKILL.md +1 -0
  170. package/bundled/skills/spec-derivation/SKILL.md +5 -1
  171. package/bundled/skills/spec-driven-dev/SKILL.md +5 -1
  172. package/bundled/skills/sweep/SKILL.md +1 -0
  173. package/bundled/skills/tdd-assistant/SKILL.md +5 -1
  174. package/bundled/skills/test-coverage-assistant/SKILL.md +5 -1
  175. package/bundled/skills/testing-guide/SKILL.md +1 -0
  176. package/package.json +1 -1
  177. package/standards-registry.json +7 -7
@@ -3,13 +3,14 @@
3
3
 
4
4
  id: ai-response-navigation
5
5
  meta:
6
- version: "1.2.0"
6
+ version: "1.3.0"
7
7
  updated: "2026-08-17"
8
8
  source: core/ai-response-navigation.md
9
9
  description: >
10
10
  Every substantive AI response must include contextual next-step suggestions with recommended
11
- options (rules 1-6, required). Optional rules 7-9 govern the answer itself: lead with the
12
- finding, restate state across turns, no preamble.
11
+ options (rules 1-6, required). Optional rules 7-11 govern the answer itself: lead with the
12
+ finding, restate state across turns, no preamble, plain language as the subject, and a
13
+ trade-off on every option rather than only the recommended one.
13
14
 
14
15
  rules:
15
16
  - id: navigation-footer
@@ -116,6 +117,32 @@ rules:
116
117
  Does NOT apply to closers: navigation-footer still stands.
117
118
  priority: optional
118
119
 
120
+ - id: plain-language-is-the-subject
121
+ trigger: any response explaining a situation, a defect, or a system's behaviour to a human
122
+ instruction: >
123
+ Explain what happened in the words the reader would use. File paths, symbol names, line
124
+ references, command output and version strings are support: they belong after the sentence
125
+ they support, not as the sentence itself. This does NOT license omitting them — a reader who
126
+ wants to verify must be able to. It governs which of the two is the subject.
127
+ Distinct from lead-with-the-finding: that rule orders finding before evidence, this one
128
+ governs register. A response can lead with its finding and still state that finding in
129
+ vocabulary only its author holds; both leave the reader unable to act.
130
+ priority: optional
131
+
132
+ - id: every-option-carries-its-trade-off
133
+ trigger: a response asking the reader to choose between two or more courses of action
134
+ instruction: >
135
+ recommendation-marking requires marking the recommended option and giving ITS reason. This
136
+ extends that to the rest: each option states what it buys and what it costs, in its own terms.
137
+ A list where only the recommendation is argued hands the comparison back to the reader, which
138
+ is the work they asked to have done; and an option shown without its downside reads as having
139
+ none. A trade-off is not a hedge — "this may be slightly harder" is not a cost, "this rewrites
140
+ 110 files and needs a human to check the translations" is. If an option genuinely has no
141
+ downside worth stating, say so explicitly: an empty cell reads as "not analysed" and the
142
+ reader cannot tell those apart. Composes with adaptive-quantity — trade-offs make each option
143
+ costlier to read, so the 1-5 cap matters more, not less.
144
+ priority: optional
145
+
119
146
  related_standards:
120
147
  - ai-command-behavior
121
148
  - ai-instruction-standards
@@ -2,7 +2,7 @@
2
2
 
3
3
  > **Language**: English | [繁體中文](../locales/zh-TW/core/ai-response-navigation.md) | [简体中文](../locales/zh-CN/core/ai-response-navigation.md)
4
4
 
5
- **Version**: 1.2.0
5
+ **Version**: 1.3.0
6
6
  **Last Updated**: 2026-08-17
7
7
  **Applicability**: All projects using AI-assisted development
8
8
  **Scope**: universal
@@ -19,11 +19,12 @@ This standard defines navigation behavior for AI responses: every substantive AI
19
19
 
20
20
  **Solution**: A standard "Navigation Footer" appended to every substantive AI response, with contextual templates, recommendation marking, and adaptive option quantities.
21
21
 
22
- **Scope note (v1.2.0)**: Rules 1–6 govern what comes *after* the answer. Rules 7–9 — added in 1.2.0
23
- and **optional** — govern the answer itself: lead with the finding, restate state across turns, no
24
- preamble. They were added because a response can satisfy every one of Rules 1–6 while burying its
25
- conclusion, and a reader who cannot find the answer is not helped by a correct footer telling them
26
- what to do next.
22
+ **Scope note (v1.2.0, extended in 1.3.0)**: Rules 1–6 govern what comes *after* the answer.
23
+ Rules 7–11 — **all optional** — govern the answer itself: lead with the finding (R7), restate state
24
+ across turns (R8), no preamble (R9), plain language as the subject (R10), and a trade-off on every
25
+ option rather than only the recommended one (R11). They exist because a response can satisfy every
26
+ one of Rules 1–6 while burying its conclusion, stating it in vocabulary only its author holds, or
27
+ listing options the reader still has to compare themselves.
27
28
 
28
29
  ---
29
30
 
@@ -103,9 +104,11 @@ Tier names are **vendor-neutral**. Each tool or platform maps these tiers to its
103
104
 
104
105
  ---
105
106
 
106
- ## The Answer Before the Navigation (Rules 7–9, Optional)
107
+ ## The Answer Before the Navigation (Rules 7–11, Optional)
107
108
 
108
- > **Borrowed from**: [`ayghri/i-have-adhd`](https://github.com/ayghri/i-have-adhd) (MIT), 3 of its 10 rules.
109
+ > **R7–R9 borrowed from**: [`ayghri/i-have-adhd`](https://github.com/ayghri/i-have-adhd) (MIT), 3 of its 10 rules.
110
+ > **R10–R11 added in 1.3.0** from a different source — a user telling the author, twice in one session,
111
+ > that a correct and complete answer was unreadable. R7–R9 had already shipped and were being followed.
109
112
  > The other 7 were dropped: 2 are already covered by Rules 1–2 above, and 5 either conflict with
110
113
  > this standard (its "no recap / no closers" contradicts Rule 1's Navigation Footer; its
111
114
  > "cap lists at 5" would truncate evidence tables and traversal denominators) or duplicate
@@ -116,7 +119,7 @@ itself — a response could bury its conclusion under a wall of evidence and sti
116
119
  in this standard by appending a correct Navigation Footer. A reader who cannot find the answer is
117
120
  not helped by being told what to do next.
118
121
 
119
- **These three rules are optional**, in the same sense as Rule 6: adopting projects are not required
122
+ **These five rules are optional**, in the same sense as Rule 6: adopting projects are not required
120
123
  to enable them, and existing skills need no retroactive update. A project MAY promote any of them to
121
124
  required in its own configuration. What is *not* optional is that they have precise triggers — a rule
122
125
  phrased so loosely that it never fires is indistinguishable from not having the rule.
@@ -164,6 +167,53 @@ but the delay it puts between the reader and the answer.
164
167
  **Rule 9 does not apply to closers.** Rule 1 requires a Navigation Footer, and that requirement
165
168
  stands — the end of a response is where this standard puts the reader's next move.
166
169
 
170
+ ### Rule 10: Plain Language Is the Subject; Identifiers Are Support (Optional)
171
+
172
+ **Trigger**: any response explaining a situation, a defect, or a system's behaviour to a human.
173
+
174
+ Explain what happened in the words the reader would use. File paths, symbol names, line
175
+ references, command output and version strings are **support** — they belong after the sentence
176
+ they support, not as the sentence itself.
177
+
178
+ | Instead of | Write |
179
+ |-----------|-------|
180
+ | "`load_topics()` at `intel-scout.py:883` reads `intel-topics.yaml`, while `load_feeds()` at `:1033` reads `intel-feeds.yaml`." | "It gathers material two ways: by searching keywords, and by subscribing to a fixed set of blogs." *(then cite both call sites)* |
181
+ | "`manifest.skillHashes` has 137 entries keyed by `claude-code/project/<n>/SKILL.md` while the lookup uses `.claude/skills/<n>`." | "The hashes are stored under one naming scheme and looked up under another, so none are ever found." *(then show both keys)* |
182
+
183
+ This is **not** a licence to omit the identifiers — a reader who wants to verify must be able to.
184
+ It governs which of the two is the subject of the sentence.
185
+
186
+ **Why it is separate from R7**: R7 orders *finding before evidence*. R10 governs *register* — a
187
+ response can lead with its finding and still state that finding in vocabulary only its author
188
+ holds. Both failures leave the reader unable to act; they are different failures.
189
+
190
+ ### Rule 11: Every Option Carries Its Own Trade-off (Optional)
191
+
192
+ **Trigger**: a response that asks the reader to choose between two or more courses of action.
193
+
194
+ Rule 2 requires marking the recommended option and giving *its* reason. Rule 11 extends that to
195
+ the rest: **each** option states what it buys and what it costs, in its own terms.
196
+
197
+ A list of options where only the recommended one is argued hands the comparison back to the
198
+ reader — which is the work they asked to have done. And an option presented without its downside
199
+ reads as having none, which is rarely true and never verifiable from the list alone.
200
+
201
+ ```markdown
202
+ > **Please choose:**
203
+ > | Option | Buys you | Costs you |
204
+ > |---|---|---|
205
+ > | **(A) …** ⭐ **Recommended** — [why this one] | … | … |
206
+ > | **(B) …** | … | … |
207
+ ```
208
+
209
+ **A trade-off is not a hedge.** "This may be slightly harder" is not a cost; "this rewrites 110
210
+ files and needs a human to check the translations" is. If an option genuinely has no downside
211
+ worth stating, say so explicitly rather than leaving the column empty — an empty cell reads as
212
+ "not analysed", and the reader cannot tell those apart.
213
+
214
+ **Composes with Rule 4**: the option count stays bounded (1–5). Trade-offs make each option
215
+ costlier to read, so this rule makes Rule 4's cap matter more, not less.
216
+
167
217
  ---
168
218
 
169
219
  ## Contextual Templates
@@ -356,6 +406,8 @@ Each tool's integration layer is responsible for rendering the Navigation Footer
356
406
  | R7 | *(Optional)* Lead with the finding; evidence follows the claim it supports |
357
407
  | R8 | *(Optional)* 3+ turns or 3+ steps → restate state in one line |
358
408
  | R9 | *(Optional)* No preamble. Closers still required — see R1 |
409
+ | R10 | *(Optional)* Plain language is the subject; identifiers support it, after the claim |
410
+ | R11 | *(Optional)* Every option states what it buys and costs — not only the recommended one |
359
411
 
360
412
  | Exempt | Not Exempt |
361
413
  |--------|------------|
@@ -379,6 +431,7 @@ Each tool's integration layer is responsible for rendering the Navigation Footer
379
431
 
380
432
  | Version | Date | Changes |
381
433
  |---------|------|---------|
434
+ | 1.3.0 | 2026-08-17 | Add optional R10–R11. R10 governs register: plain language is the subject of the sentence and identifiers support it — distinct from R7, which orders finding before evidence, because a response can lead with its finding and still state it in vocabulary only its author holds. R11 extends Rule 2 from the recommended option to all of them: a list where only the recommendation is argued hands the comparison back to the reader, and an option shown without its cost reads as having none |
382
435
  | 1.2.0 | 2026-08-17 | Add optional R7–R9 governing the answer itself (lead with the finding, restate state, no preamble). Borrowed from `ayghri/i-have-adhd` (MIT), 3 of its 10 rules; the other 7 were dropped as duplicated by R1–R2, in conflict with R1, or covered by estimation-standards. Rules 1–6 could all be satisfied by a response that buries its conclusion — R7–R9 close that |
383
436
  | 1.1.0 | 2026-06-10 | Add R6 optional model tier annotation (`〔model: Fast\|Standard\|Capable〕`); vendor-neutral; no forced changes to existing skills |
384
437
  | 1.0.0 | 2026-03-25 | Initial release |
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  source: ../../CHANGELOG.md
3
- source_version: 6.6.0
4
- translation_version: 6.6.0
3
+ source_version: 6.7.0
4
+ translation_version: 6.7.0
5
5
  last_synced: 2026-08-17
6
6
  status: current
7
7
  ---
@@ -17,10 +17,29 @@ status: current
17
17
 
18
18
  ## [Unreleased]
19
19
 
20
+ ## [6.7.0] - 2026-08-17
21
+
22
+ ### 修正
23
+
24
+ - **skill 触发面在三层全数复原——55 个 skill × 英文、zh-TW、zh-CN。** 一个 skill 的 `description` 是模型决定要不要调用它时**唯一看得到的东西**。commit `d415937e`(2026-02-10)把其中 17 个改写成 `[UDS] <标签>`,删掉 `Use when:` 与 `Keywords:` 两行、**连中文关键字一起**;相邻 commit 的标题写着 `token optimization`。当时**只有 token 可数**——没有任何东西在量触发面,于是那个取舍看起来是单边的,而它不是。2026-08-14 测量:55 个含 `SKILL.md` 的 skill 中,**27 个有触发条件、27 个有关键字、0 个有排除条件、28 个两者皆无**——而那 28 个正是方法论核心:tdd、bdd、atdd、spec-driven-dev、code-review、commit-standards、checkin、requirement。
25
+ - **英文来源**:28 个复原,且**全部 55 个补上排除条件**(`Not for:`)。只加触发不加排除,换来的是过度触发;而一个不该响却响的 skill 会被整个关掉,**连带拖走还能用的那些**。15 个可从 git 历史救回者中**有 8 个被重写**,因为历史文字已不描述现行行为——其中六个宣称自己会引导某套生命周期,而那已于 XSPEC-095 移交采用层。**救回一个过期的描述,比不救更糟。**
26
+ - **语系层**:zh-TW 与 zh-CN **各 55/55,是翻译不是转码**。**这一半才是重点**——以 `--locale zh-tw` 安装的项目,在英文来源已修好时仍拿到被剥过的描述,这个修正本来到不了它要给的那个读者。繁体与简体各自用地道用词,另修正三份 zh-CN 描述中混入的繁体字。
27
+ - **翻译 drift 62 → 38**,剩下的 38 是刻意的停点:24 份漂移完全来自本次 description 编辑者更新了 hash,12 份正文早已漂移者保留过期 hash——更新它们等于宣称整份文件已同步,而那件事没有人验证过。
28
+ - ⚠️ **28 个中有 10 个带着 `disable-model-invocation: true`**,由同一个 `d415937e` 加入。对那 10 个而言,补描述**并不会**让它们变成可被选中——挡住的是那个标志,而要不要拿掉它是**设计决定不是缺陷**。模型真正选得到的数量是 **45,不是 55**。
29
+
30
+ ### 新增
31
+
32
+ - **`ai-response-navigation` 1.2.0 → 1.3.0 —— 可选规则 R10 与 R11。** 它们的来源与 R7–R9 不同:用户在同一次工作会话中**两度**指出,一个正确且完整的回答读不懂,而当时 R7–R9 已经出货且正在被遵守。**先讲发现并不足够。**
33
+ - **R10 —— 白话是主语,标识符是佐证。** *触发*:任何向人解释一个情况、一个缺陷、或一个系统行为的响应。用读者会用的话说清楚发生了什么;路径、符号、行号、命令输出、版本字符串属于它们所支持的那句话**之后**,而不是那句话本身。它**不是**可以省略它们的许可——想验证的读者必须验得了。**与 R7 分开是刻意的**:R7 管的是「先发现后证据」的顺序,而一个响应可以先讲发现、却仍用只有作者持有的词汇讲它。两者都让读者无法行动,但它们是不同的失效。
34
+ - **R11 —— 每个选项都要带自己的利弊。** *触发*:要求读者在两个以上做法之间选择的响应。规则 2 已要求标示推荐项并给出**它的**理由;R11 要求**每一个**选项都说明它换到什么、代价是什么。一份只有推荐项被论证的清单,等于**把比较的工作丢回给读者**——而那正是他请你做的事;而没标代价的选项读起来像是没有代价。**利弊不是模棱两可**:「稍微难一点」不是代价,「重写 110 个文件且翻译需要人审」才是。空白的代价栏读起来是「没有分析过」,而读者分不出这两者。
35
+
20
36
  ## [6.6.0] - 2026-08-17
21
37
 
22
38
  ### 新增
23
39
 
40
+ - **`spec-driven-development` 2.3.0 → 2.4.0 —— 没有验证项的 AC 不是 AC。** 每一条验收标准都必须有一个**指向它的验证项**——测试、检查、闸门,或一则明确记录的手动步骤。没有任何验证项引用的 AC 是一张没有人兑现的支票,而且它**不会**大声失败:它只是安静地停止成立,而规格继续宣称它为真。**规则**:这样的 AC 必须**降级为设计意图**,不得继续挂在 AC 栏——降级是诚实的,未经验证的 AC 不是。标准同时写明什么**不算**验证项:「审查者会注意到」不是验证项,因为审查者读的是规格,而规格说那条 AC 成立。新增规则 `SDD-AC-VERIFIED`。
41
+ - **实测案例**:一份 2026-05-14 的规格写着 `AC-7:旧版系统报告完整保留(无回归)`。它的 Test Plan 有七项,**没有一项指向 AC-7**。该报告的 timer 是 disabled、部署函数从未被调用——**从那条 AC 被写下的同一天起**。三个月后在验证一件无关的安装时偶然撞到。它不是接上了检查后来松脱,**它从来没有被接上过**。
42
+
24
43
  - **`ai-response-navigation` 1.1.0 → 1.2.0 —— 可选规则 R7–R9,管答案本身**(XSPEC 借鉴 B-10,来源 [`ayghri/i-have-adhd`](https://github.com/ayghri/i-have-adhd),MIT)。规则 1–6 管的是答案**之后**要附什么:导航区块、标记过的推荐、匹配响应类型的模板。**答案本身没有任何规则在管。** 于是一个响应可以把结论埋在一整面证据底下,只要结尾附上正确的导航区块,它仍然满足**本标准的每一条**——而找不到答案的读者,不会因为被告知下一步而得到帮助。
25
44
  - **R7 —— 先讲发现,不要先讲过程。** *触发*:回答问题、汇报调查结果、或提出决策的响应。第一行写**查到了什么**或**该做什么**——不是方法、不是把问题复述一遍、不是回答的计划。证据(`file:line`、命令输出、表格、测量数字)是**佐证**,应放在它所支持的论断之后;以证据开场会迫使读者自行重建结论,而那正是他请你做的工作。本条规范的是**顺序**,**不**代表可以省略证据。
26
45
  - **R8 —— 每一轮重述进度。** *触发*:跨 3 轮以上的对话,或含 3 个以上步骤的任务。用一行说明工作进行到哪里。不能假设读者能在消息之间记住「我们在 5 步中的第 3 步」,而重述它的成本是一个句子。与模板 4(进行中)互补:**R8 管开头,模板管结尾。**
@@ -15,7 +15,7 @@ status: current
15
15
 
16
16
  > **语言**: [English](../../README.md) | [繁體中文](../zh-TW/README.md) | 简体中文
17
17
 
18
- **版本**: 6.6.0 | **发布日期**: 2026-08-17 | **授权**: [双重授权](../../LICENSE) (CC BY 4.0 + MIT)
18
+ **版本**: 6.7.0 | **发布日期**: 2026-08-17 | **授权**: [双重授权](../../LICENSE) (CC BY 4.0 + MIT)
19
19
 
20
20
  语言无关、框架无关的软件项目文档标准。通过 AI 原生工作流,确保不同技术栈之间的一致性、质量和可维护性。
21
21
 
@@ -13,7 +13,7 @@ status: current
13
13
  <!-- UDS_SUPPORTED_VERSIONS_START -->
14
14
  | 版本 | 支持状态 |
15
15
  |------|--------|
16
- | 6.6.0 | ✅ 最新正式版 |
16
+ | 6.7.0 | ✅ 最新正式版 |
17
17
  | < 6.0.0 | ❌ 已终止支持 |
18
18
  <!-- UDS_SUPPORTED_VERSIONS_END -->
19
19
 
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  source: ../../../core/ai-response-navigation.md
3
- source_version: 1.2.0
4
- translation_version: 1.2.0
3
+ source_version: 1.3.0
4
+ translation_version: 1.3.0
5
5
  last_synced: 2026-08-17
6
6
  status: current
7
7
  ---
@@ -10,7 +10,7 @@ status: current
10
10
 
11
11
  > **语言**: [English](../../../core/ai-response-navigation.md) | [繁體中文](../../zh-TW/core/ai-response-navigation.md) | 简体中文
12
12
 
13
- **版本**: 1.2.0
13
+ **版本**: 1.3.0
14
14
  **最后更新**: 2026-08-17
15
15
  **适用范围**: 所有使用 AI 辅助开发的项目
16
16
  **范围**: universal
@@ -27,8 +27,9 @@ status: current
27
27
 
28
28
  **解决方案**:在每个实质性 AI 响应结尾附加标准化的「导航区块」,包含情境模板、推荐标记和弹性选项数量。
29
29
 
30
- **范围注记(v1.2.0)**:规则 1–6 管的是答案**之后**要附什么。规则 7–9 于 1.2.0 新增、属**可选**,
31
- 管的是答案本身:先讲发现、每轮重述进度、不要开场白。新增的理由是——
30
+ **范围注记(v1.2.0,v1.3.0 扩充)**:规则 1–6 管的是答案**之后**要附什么。
31
+ 规则 7–11 **全部属可选**,管的是答案本身:先讲发现(R7)、每轮重述进度(R8)、不要开场白(R9)、
32
+ 白话是主语(R10)、每个选项都要带自己的利弊而非只有推荐项有(R11)。新增的理由是——
32
33
  一个响应可以满足规则 1–6 的每一条,同时把结论埋起来;
33
34
  **而找不到答案的读者,不会因为结尾有一个正确的导航区块被告知下一步而得到帮助。**
34
35
 
@@ -108,9 +109,11 @@ status: current
108
109
 
109
110
  ---
110
111
 
111
- ## 导航之前的那个答案(规则 7–9,可选)
112
+ ## 导航之前的那个答案(规则 7–11,可选)
112
113
 
113
- > **借鉴自**:[`ayghri/i-have-adhd`](https://github.com/ayghri/i-have-adhd)(MIT),十条中取三条。
114
+ > **规则 7–9 借鉴自**:[`ayghri/i-have-adhd`](https://github.com/ayghri/i-have-adhd)(MIT),十条中取三条。
115
+ > **规则 10–11 于 1.3.0 新增**,来源不同——用户在同一次工作会话中两度指出,
116
+ > 一个正确且完整的回答读不懂。当时规则 7–9 已经出货且正在被遵守。
114
117
  > 其余七条删去:两条已被上方规则 1–2 涵盖,五条与本标准冲突
115
118
  > (它的「不要 recap/不要结语」与规则 1 的导航区块直接矛盾;它的「列表上限 5 项」
116
119
  > 会截断证据表格与遍历分母)或与 [estimation-standards](estimation-standards.md) 重复。
@@ -119,7 +122,7 @@ status: current
119
122
  一个响应可以把结论埋在一整面证据底下,只要结尾附上正确的导航区块,
120
123
  它仍然满足本标准的每一条。**找不到答案的读者,不会因为被告知下一步而得到帮助。**
121
124
 
122
- **这三条是可选的**,语义同规则 6:采用项目不必启用,既有 skill 也不需回头补。
125
+ **这五条是可选的**,语义同规则 6:采用项目不必启用,既有 skill 也不需回头补。
123
126
  项目**可以**在自己的配置中把任一条提升为必须。**不可选的是它们必须有精确的触发条件**——
124
127
  一条松到永远不会启动的规则,与没有这条规则无从分辨。
125
128
 
@@ -161,6 +164,46 @@ status: current
161
164
  **规则 9 不适用于结语。** 规则 1 要求的导航区块依然成立——
162
165
  响应的结尾正是本标准安放「读者下一步」的位置。
163
166
 
167
+ ### 规则 10:白话是主语,标识符是佐证(可选)
168
+
169
+ **触发条件**:任何向人解释一个情况、一个缺陷、或一个系统行为的响应。
170
+
171
+ 用读者会用的话说清楚发生了什么。文件路径、符号名称、行号、命令输出、版本字符串是**佐证**——
172
+ 它们属于它们所支持的那句话**之后**,而不是那句话本身。
173
+
174
+ | 不要写 | 改写成 |
175
+ |---|---|
176
+ | 「`intel-scout.py:883` 的 `load_topics()` 读 `intel-topics.yaml`,`:1033` 的 `load_feeds()` 读 `intel-feeds.yaml`。」 | 「它有两种取材方式:搜关键字,和订阅固定几个博客。」*(然后再附两个调用点)* |
177
+ | 「`manifest.skillHashes` 有 137 条,key 是 `claude-code/project/<n>/SKILL.md`,而查询用的是 `.claude/skills/<n>`。」 | 「哈希值存在一套命名下,查询用的是另一套,所以一个都找不到。」*(然后再贴两边的 key)* |
178
+
179
+ 这**不是**可以省略标识符的许可——想验证的读者必须验得了。它规范的是**两者之中哪一个当主语**。
180
+
181
+ **为什么它与 R7 是两条**:R7 规范的是**先讲发现再给证据**的顺序。R10 规范的是**语域**——
182
+ 一个响应可以先讲发现,却仍然用只有作者持有的词汇讲那个发现。两者都让读者无法行动,但它们是不同的失效。
183
+
184
+ ### 规则 11:每个选项都要带自己的利弊(可选)
185
+
186
+ **触发条件**:要求读者在两个以上做法之间选择的响应。
187
+
188
+ 规则 2 要求标示推荐选项并给出**它的**理由。规则 11 把这件事扩及其余:
189
+ **每一个**选项都要用它自己的话说明它换到什么、代价是什么。
190
+
191
+ 一份只有推荐项被论证的选项清单,等于**把比较的工作丢回给读者**——而那正是他请你做的事。
192
+ 而一个没有标出代价的选项,读起来像是没有代价,这很少为真,且从清单本身无从验证。
193
+
194
+ ```markdown
195
+ > **请选择:**
196
+ > | 选项 | 换到什么 | 代价是什么 |
197
+ > |---|---|---|
198
+ > | **(A) …** ⭐ **推荐** — [为什么是这个] | … | … |
199
+ > | **(B) …** | … | … |
200
+ ```
201
+
202
+ **利弊不是模棱两可。**「这可能会稍微难一点」不是代价;「这要重写 110 个文件,而且翻译需要人审」才是。
203
+ 若某个选项真的没有值得一提的代价,**明说**,不要留空——空白读起来是「没有分析过」,而读者分不出这两者。
204
+
205
+ **与规则 4 相辅**:选项数维持在 1–5。利弊让每个选项读起来更花力气,所以这条规则让规则 4 的上限**更**要紧,不是更不要紧。
206
+
164
207
  ---
165
208
 
166
209
  ## 情境模板
@@ -353,6 +396,8 @@ AI 需要用户做出选择或提供信息时使用。
353
396
  | R7 | *(可选)* 先讲发现;证据放在它所支持的论断之后 |
354
397
  | R8 | *(可选)* 跨 3 轮或 3 步以上 → 用一行重述进度 |
355
398
  | R9 | *(可选)* 不要开场白。结语仍为必须——见 R1 |
399
+ | R10 | *(可选)* 白话是主语;标识符放在论断之后当佐证 |
400
+ | R11 | *(可选)* 每个选项都要说明换到什么、代价是什么——不只推荐那一个 |
356
401
 
357
402
  | 豁免 | 不豁免 |
358
403
  |------|--------|
@@ -376,6 +421,7 @@ AI 需要用户做出选择或提供信息时使用。
376
421
 
377
422
  | 版本 | 日期 | 变更 |
378
423
  |------|------|------|
424
+ | 1.3.0 | 2026-08-17 | 新增可选规则 R10–R11。R10 管语域:白话是句子的主语、标识符当佐证——与 R7 不同,R7 管的是「先发现后证据」的顺序,而一个响应可以先讲发现却仍用只有作者持有的词汇讲它。R11 把规则 2 从推荐选项扩及全部:只论证推荐项的清单等于把比较丢回给读者,而没标代价的选项读起来像没有代价 |
379
425
  | 1.2.0 | 2026-08-17 | 新增可选规则 R7–R9,管答案本身(先讲发现、重述进度、不要开场白)。借鉴自 `ayghri/i-have-adhd`(MIT),十条取三;其余七条因已被 R1–R2 涵盖、与 R1 冲突、或与 estimation-standards 重复而删去。规则 1–6 全部可以被一个把结论埋起来的响应满足——R7–R9 补上这个缺口 |
380
426
  | 1.1.0 | 2026-06-10 | 新增规则 R6 可选模型级别标注(`〔模型:Fast|Standard|Capable〕`);与厂商无关;不强制既有技能回改 |
381
427
  | 1.0.0 | 2026-03-25 | 初始版本 |
@@ -2,12 +2,16 @@
2
2
  name: ac-coverage
3
3
  source: ../../../../skills/ac-coverage/SKILL.md
4
4
  source_version: 1.1.0
5
- translation_version: 1.1.0
6
- last_synced: 2026-07-09
7
- source_hash: eac0755fd844
5
+ translation_version: 1.2.0
6
+ last_synced: 2026-08-17
7
+ source_hash: 168298f3a994
8
8
  scope: universal
9
9
  status: current
10
- description: "[UDS] 分析验收条件(AC)与测试之间的追踪关系并生成覆盖率报告"
10
+ description: |
11
+ [UDS] 分析验收条件(AC)与测试之间的追踪关系,并生成需求层级的覆盖率报告。
12
+ Use when: 审计哪些验收条件已有测试、从 SPEC 文件建立追踪矩阵、发布前找出尚未被覆盖的 AC。
13
+ Not for: 代码层级的行/分支/函数覆盖率——请用 /coverage;补写缺少的测试——请用 /tdd 或 /spec-derive。
14
+ Keywords: AC coverage, traceability, acceptance criteria, SPEC, traceability matrix, 验收条件, 需求追踪, 覆盖率矩阵, 追踪矩阵.
11
15
  allowed-tools: Read, Grep, Glob
12
16
  argument-hint: "[规格文件路径]"
13
17
  ---
@@ -1,14 +1,15 @@
1
1
  ---
2
2
  source: ../../../../skills/adr-assistant/SKILL.md
3
3
  source_version: 1.0.0
4
- source_hash: 900a9a7b5b04
5
- translation_version: 1.0.0
6
- last_synced: 2026-07-09
4
+ source_hash: a5c79c6a04fe
5
+ translation_version: 1.1.0
6
+ last_synced: 2026-08-17
7
7
  status: current
8
8
  description: |
9
- [UDS] 创建、管理和追踪架构决策记录(ADR)。
10
- 使用时机:架构决策、技术选型、设计取舍、模式选择。
11
- 关键字:ADR, architecture decision, decision record, 架构决策, 决策记录。
9
+ [UDS] 创建、管理并追踪架构决策记录(ADR)。
10
+ Use when: 架构决策、技术选型、设计取舍、模式选择。
11
+ Not for: 不改变架构的决策——记在规格或 commit 里即可;想法还没成形到足以下决定——请用 /brainstorm。
12
+ Keywords: ADR, architecture decision, decision record, trade-off, 架构决策, 决策记录, 设计取舍.
12
13
  ---
13
14
 
14
15
  # 架构决策记录助手
@@ -1,9 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/ai-collaboration-standards/SKILL.md
3
3
  source_version: 1.1.0
4
- translation_version: 1.1.0
5
- last_synced: 2026-01-25
4
+ translation_version: 1.2.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
+ description: |
8
+ 防止 AI 幻觉,确保分析代码或提出建议时给出以证据为基础的回应。
9
+ Use when: 分析代码、提出建议、提供选项,或用户询问把握度/确定性时。
10
+ Not for: 编写 AI 指令文件本身——请用 /ai-instruction-standards;审查具体的 diff——请用 /code-review。
11
+ Keywords: certainty, assumption, inference, evidence, source, 证据, 假设, 推论, 确定性, 反幻觉.
7
12
  ---
8
13
 
9
14
  # AI 协作标准
@@ -1,14 +1,15 @@
1
1
  ---
2
2
  source: ../../../../skills/ai-friendly-architecture/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-02-05
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  scope: universal
8
8
  description: |
9
- 设计 AI 友善架构,包含明确模式、分层文档和语义边界。
10
- 使用时机:为 AI 协作规划项目结构、优化代码库以利 AI 分析、设置 AI 上下文。
11
- 关键字:architecture, AI-friendly, context, modules, documentation layers, .ai-context.yaml, 架构, AI 友善, 上下文.
9
+ 设计 AI 友善架构,包含明确的模式、分层文档与语义边界。
10
+ Use when: 为 AI 协作规划项目结构、优化代码库以利 AI 分析、建立 AI 上下文。
11
+ Not for: 编写指令文件本身——请用 /ai-instruction-standards;依语言惯例安排目录布局——请用 /project-structure-guide。
12
+ Keywords: architecture, AI-friendly, context, modules, documentation layers, .ai-context.yaml, 架构, AI 友善, 上下文, 语义边界.
12
13
  ---
13
14
 
14
15
  # AI 友善架构指南
@@ -1,14 +1,15 @@
1
1
  ---
2
2
  source: ../../../../skills/ai-instruction-standards/SKILL.md
3
3
  source_version: 2.0.0
4
- translation_version: 2.0.0
5
- last_synced: 2026-05-05
4
+ translation_version: 2.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  scope: partial
8
8
  description: |
9
- 创建和维护 AI 指令文件(CLAUDE.md、.cursorrules 等)并采用适当结构。
10
- 使用时机:创建 AI 指令文件、区分通用与项目特定规则、配置 AI 工具。
11
- 关键字:CLAUDE.md, cursorrules, windsurfrules, clinerules, AI instructions, system prompt, 指令文件, AI 设定.
9
+ 创建并维护 AI 指令文件(CLAUDE.md、AGENTS.md、.cursor/rules/ 等),并采用适当结构。
10
+ Use when: 创建 AI 指令文件、区分通用规则与项目特定规则、配置 AI 工具。
11
+ Not for: 调整代码库结构让 AI 好导览——请用 /ai-friendly-architecture;要求以证据为基础的回答——请用 /ai-collaboration-standards。
12
+ Keywords: CLAUDE.md, AGENTS.md, cursorrules, windsurfrules, clinerules, AI instructions, system prompt, 指令文件, AI 设定, 系统提示词.
12
13
  ---
13
14
 
14
15
  # AI 指令文件标准指南
@@ -1,13 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/api-design-assistant/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-03-23
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  description: |
8
- 引导 API 设计,遵循 REST、GraphQL 和 gRPC 最佳实践。
9
- 使用时机:设计 API、审查端点、API 版本决策。
10
- 关键字:API, REST, GraphQL, gRPC, endpoint, versioning, 接口设计。
8
+ 引导 API 设计,遵循 REST、GraphQL 与 gRPC 最佳实践。
9
+ Use when: 设计 API、审查端点、API 版本策略决策。
10
+ Not for: 验证运行中的 API 是否符合消费端期待——请用 /contract-test;API 背后的 schema 设计——请用 /database。
11
+ Keywords: API, REST, GraphQL, gRPC, endpoint, versioning, 接口设计, 端点, 版本策略.
11
12
  ---
12
13
 
13
14
  # API 设计助手
@@ -1,13 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/atdd-assistant/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-02-10
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  description: |
8
- 引导验收测试驱动开发(ATDD)流程,用于定义和验证用户故事。
9
- 使用时机:定义验收条件、用户故事验证、ATDD 流程。
10
- 关键字:atdd, acceptance, user story, INVEST, 验收测试, 用户故事, 验收条件。
8
+ [UDS] 验收测试驱动开发(ATDD)的参考资料:INVEST 准则、Gherkin 验收条件格式与 Three Amigos 结构。
9
+ Use when: 与 Product Owner 一起定义验收条件、进行规格工作坊、用 INVEST 检视用户故事。
10
+ Not for: 执行 ATDD 生命周期或强制 PO 签核关卡——该部分已移至采用层(XSPEC-095);编写单元测试——请用 /tdd。
11
+ Keywords: ATDD, acceptance test, acceptance criteria, INVEST, specification workshop, Three Amigos, 验收测试驱动开发, 验收条件, 规格工作坊.
11
12
  ---
12
13
 
13
14
  # ATDD 助手
@@ -1,13 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/audit-assistant/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-03-04
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  description: |
8
- UDS 健康检查与反馈系统。诊断安装完整性、检测开发模式、识别摩擦点。
9
- 使用时机:检查 UDS 安装、报告问题、发现规范化机会。
10
- 关键字:audit, health, pattern, friction, feedback, 审计, 健康, 模式, 摩擦, 反馈。
8
+ [UDS] 诊断 UDS 安装的健康状态,并向上游提交结构化反馈。
9
+ Use when: .standards/ 看起来损坏或不同步、验证 manifest 完整性、反馈既有 UDS 标准用起来的摩擦点。
10
+ Not for: 审计你自己应用程序的代码质量——请用 /metrics 或 /code-review;依赖包与密钥扫描——请用 /scan。
11
+ Keywords: UDS audit, health check, manifest integrity, standards feedback, friction, 安装健康, 标准审计, 反馈, 完整性检查.
11
12
  ---
12
13
 
13
14
  # 审计助手
@@ -1,13 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/bdd-assistant/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-02-10
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  description: |
8
- 引导行为驱动开发(BDD)流程,使用 Given-When-Then 格式撰写场景。
9
- 使用时机:定义行为场景、撰写 Gherkin、三剑客会议、BDD 流程。
10
- 关键字:bdd, gherkin, given-when-then, scenario, feature, 行为驱动, 场景, 验收。
8
+ [UDS] 行为驱动开发(BDD)的参考资料:Gherkin 的 Given-When-Then 格式与 Three Amigos 结构。
9
+ Use when: 编写或审查 .feature 场景、选定通用语言、针对行为进行探索式对话。
10
+ Not for: 执行 BDD 生命周期或 RED/GREEN 自动化——该部分已移至采用层(XSPEC-095);把 .feature 文件转成 E2E 骨架——请用 /e2e。
11
+ Keywords: BDD, Gherkin, Given When Then, feature file, scenario, Three Amigos, 行为驱动开发, 场景, 特性文件, 通用语言.
11
12
  ---
12
13
 
13
14
  # BDD 助手
@@ -2,14 +2,15 @@
2
2
  name: brainstorm-assistant
3
3
  source: ../../../../skills/brainstorm-assistant/SKILL.md
4
4
  source_version: 4.1.0
5
- source_hash: 0c1fa3392232
6
- translation_version: 4.1.0
7
- last_synced: 2026-07-09
5
+ source_hash: 711404f2c4af
6
+ translation_version: 4.2.0
7
+ last_synced: 2026-08-17
8
8
  status: current
9
9
  description: |
10
- 在编写规格前进行结构化 AI 辅助头脑风暴。
11
- 使用时机:功能规划、创意发想、问题定义。
12
- 关键字:brainstorm, persona, multi-critic, HMW, SCAMPER, 头脑风暴, 发想。
10
+ [UDS] 在规格出现之前执行的结构化多角色头脑风暴,并附带评分质量关卡。
11
+ Use when: 想法还很模糊、在敲定方向前探索替代方案、需要多样性而不是第一个看似合理的答案。
12
+ Not for: 方向已经定了的工作规划——请用 /plan 或 /sdd;记录已经做成的决策——请用 /adr。
13
+ Keywords: brainstorm, ideation, divergence, convergence, persona ensemble, devil advocate, 头脑风暴, 发想, 发散收敛, 多角色.
13
14
  ---
14
15
 
15
16
  # 头脑风暴助手
@@ -1,13 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/changelog-guide/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-02-10
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  description: |
8
- 根据 Keep a Changelog 格式产生和维护 CHANGELOG.md 条目。
9
- 使用时机:更新变更日志、版本发布、记录变更。
10
- 关键字:changelog, CHANGELOG.md, keep a changelog, 变更日志, 版本记录。
8
+ [UDS] 以 Keep a Changelog 格式生成并维护 CHANGELOG.md 条目。
9
+ Use when: 从 commit 历史撰写变更日志条目、填写 Unreleased 区段、将变更分类为 Added/Changed/Fixed。
10
+ Not for: 决定下一个版本号或执行发版——请用 /release;撰写 commit message 本身——请用 /commit。
11
+ Keywords: changelog, CHANGELOG.md, Keep a Changelog, release notes, unreleased, 变更日志, 发布说明, 版本记录.
11
12
  ---
12
13
 
13
14
  # 变更日志助手
@@ -1,13 +1,14 @@
1
1
  ---
2
2
  source: ../../../../skills/checkin-assistant/SKILL.md
3
3
  source_version: 1.0.0
4
- translation_version: 1.0.0
5
- last_synced: 2026-02-10
4
+ translation_version: 1.1.0
5
+ last_synced: 2026-08-17
6
6
  status: current
7
7
  description: |
8
- 在提交代码前验证品质关卡,确保代码库稳定性。
9
- 使用时机:提交前检查、品质验证、pre-commit 检查。
10
- 关键字:checkin, pre-commit, quality gate, 签入, 品质关卡, 验证。
8
+ [UDS] 提交前质量关卡的参考资料:关卡定义、检查清单项目,以及绝不可提交的规则。
9
+ Use when: 决定 commit 前必须通过哪些检查、审计项目实际强制了哪些质量关卡、确认是否已可签入。
10
+ Not for: 执行关卡流程或中止 commit——该部分已移至采用层(XSPEC-095);找出并清除调试残留——请用 /sweep。
11
+ Keywords: check-in, pre-commit, quality gate, commit readiness, never commit, 签入, 提交前检查, 质量关卡, 检查清单.
11
12
  ---
12
13
 
13
14
  # 签入助手