tribunal-kit 8.0.0 → 9.0.0

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 (258) hide show
  1. package/.agent/ARCHITECTURE.md +2 -2
  2. package/.agent/config/plugin.json +4 -4
  3. package/.agent/config/system-prompt.md +54 -1
  4. package/.agent/routing_index.json +2 -2
  5. package/.agent/skills/12-principles-of-animation/SKILL.md +19 -16
  6. package/.agent/skills/60fps-animation/SKILL.md +47 -21
  7. package/.agent/skills/accessible-animation/SKILL.md +19 -16
  8. package/.agent/skills/adapt/SKILL.md +19 -16
  9. package/.agent/skills/advanced-rag-pipelines/SKILL.md +41 -40
  10. package/.agent/skills/agent-organizer/SKILL.md +9 -59
  11. package/.agent/skills/agentic-patterns/SKILL.md +9 -57
  12. package/.agent/skills/agentic-workflows-2026/SKILL.md +45 -8
  13. package/.agent/skills/ai-app-hardening/SKILL.md +31 -5
  14. package/.agent/skills/ai-prompt-injection-defense/SKILL.md +28 -66
  15. package/.agent/skills/animation-on-scroll/SKILL.md +19 -16
  16. package/.agent/skills/animation-systems/SKILL.md +19 -15
  17. package/.agent/skills/antfu-conventions/SKILL.md +19 -18
  18. package/.agent/skills/api-patterns/SKILL.md +8 -58
  19. package/.agent/skills/api-security-auditor/SKILL.md +8 -58
  20. package/.agent/skills/app-builder/SKILL.md +9 -59
  21. package/.agent/skills/appflow-wireframe/SKILL.md +13 -57
  22. package/.agent/skills/apple-design/SKILL.md +22 -14
  23. package/.agent/skills/architecture/SKILL.md +9 -59
  24. package/.agent/skills/audit-and-fix/SKILL.md +19 -16
  25. package/.agent/skills/authentication-best-practices/SKILL.md +8 -58
  26. package/.agent/skills/backend-security-expert/SKILL.md +32 -65
  27. package/.agent/skills/baseline-ui/SKILL.md +22 -14
  28. package/.agent/skills/bash-linux/SKILL.md +9 -59
  29. package/.agent/skills/behavioral-modes/SKILL.md +9 -57
  30. package/.agent/skills/better-colors/SKILL.md +22 -14
  31. package/.agent/skills/better-typography/SKILL.md +26 -14
  32. package/.agent/skills/better-ui/SKILL.md +54 -19
  33. package/.agent/skills/bolder/SKILL.md +19 -16
  34. package/.agent/skills/brainstorming/SKILL.md +9 -57
  35. package/.agent/skills/browser-native-ai/SKILL.md +9 -28
  36. package/.agent/skills/build-primitive/SKILL.md +19 -16
  37. package/.agent/skills/building-native-ui/SKILL.md +9 -59
  38. package/.agent/skills/cicd-pro/SKILL.md +21 -28
  39. package/.agent/skills/clarify/SKILL.md +19 -16
  40. package/.agent/skills/clean-code/SKILL.md +18 -61
  41. package/.agent/skills/cloud-architect/SKILL.md +21 -28
  42. package/.agent/skills/cobejs/SKILL.md +19 -15
  43. package/.agent/skills/code-review-checklist/SKILL.md +8 -58
  44. package/.agent/skills/codebase-design/SKILL.md +33 -22
  45. package/.agent/skills/colorize/SKILL.md +19 -16
  46. package/.agent/skills/compact-landing/SKILL.md +19 -16
  47. package/.agent/skills/company-logos/SKILL.md +19 -16
  48. package/.agent/skills/config-validator/SKILL.md +9 -59
  49. package/.agent/skills/containerization-pro/SKILL.md +21 -28
  50. package/.agent/skills/context-engineering-pro/SKILL.md +31 -5
  51. package/.agent/skills/create-design-md/SKILL.md +22 -14
  52. package/.agent/skills/critique/SKILL.md +20 -11
  53. package/.agent/skills/csharp-developer/SKILL.md +8 -58
  54. package/.agent/skills/data-validation-schemas/SKILL.md +9 -59
  55. package/.agent/skills/database-design/SKILL.md +27 -67
  56. package/.agent/skills/delight/SKILL.md +19 -16
  57. package/.agent/skills/deployment-procedures/SKILL.md +9 -59
  58. package/.agent/skills/design-lab/SKILL.md +19 -16
  59. package/.agent/skills/devops-engineer/SKILL.md +8 -58
  60. package/.agent/skills/devops-incident-responder/SKILL.md +9 -59
  61. package/.agent/skills/diagnosing-bugs/SKILL.md +20 -21
  62. package/.agent/skills/distill/SKILL.md +19 -16
  63. package/.agent/skills/documentation-templates/SKILL.md +9 -59
  64. package/.agent/skills/domain-modeling/SKILL.md +19 -18
  65. package/.agent/skills/duckdb-analytical-sql/SKILL.md +31 -5
  66. package/.agent/skills/edge-ai-mobile/SKILL.md +31 -5
  67. package/.agent/skills/edge-computing/SKILL.md +9 -59
  68. package/.agent/skills/emil-design-eng/SKILL.md +21 -24
  69. package/.agent/skills/error-resilience/SKILL.md +31 -66
  70. package/.agent/skills/expo-router-v4/SKILL.md +31 -5
  71. package/.agent/skills/extract-design-system/SKILL.md +9 -59
  72. package/.agent/skills/fabel-protocol/SKILL.md +26 -8
  73. package/.agent/skills/fixing-accessibility/SKILL.md +22 -14
  74. package/.agent/skills/fixing-metadata/SKILL.md +19 -16
  75. package/.agent/skills/fixing-motion-performance/SKILL.md +22 -14
  76. package/.agent/skills/framer-motion-expert/SKILL.md +9 -57
  77. package/.agent/skills/frontend-design/SKILL.md +27 -7
  78. package/.agent/skills/frontend-security-expert/SKILL.md +8 -58
  79. package/.agent/skills/game-design-expert/SKILL.md +9 -59
  80. package/.agent/skills/game-engineering-expert/SKILL.md +9 -59
  81. package/.agent/skills/generative-ui-expert/SKILL.md +9 -28
  82. package/.agent/skills/geo-fundamentals/SKILL.md +9 -59
  83. package/.agent/skills/git-pro/SKILL.md +19 -30
  84. package/.agent/skills/github-operations/SKILL.md +8 -58
  85. package/.agent/skills/gpt-taste/SKILL.md +19 -16
  86. package/.agent/skills/gsap-core/SKILL.md +8 -78
  87. package/.agent/skills/gsap-frameworks/SKILL.md +8 -78
  88. package/.agent/skills/gsap-performance/SKILL.md +9 -77
  89. package/.agent/skills/gsap-plugins/SKILL.md +9 -77
  90. package/.agent/skills/gsap-react/SKILL.md +9 -21
  91. package/.agent/skills/gsap-scrolltrigger/SKILL.md +8 -78
  92. package/.agent/skills/gsap-timeline/SKILL.md +9 -77
  93. package/.agent/skills/gsap-utils/SKILL.md +9 -77
  94. package/.agent/skills/harden/SKILL.md +19 -16
  95. package/.agent/skills/harness-protocol/SKILL.md +9 -21
  96. package/.agent/skills/i18n-localization/SKILL.md +9 -59
  97. package/.agent/skills/impeccable/SKILL.md +19 -18
  98. package/.agent/skills/improve-codebase-architecture/SKILL.md +19 -17
  99. package/.agent/skills/improve-ui/SKILL.md +22 -14
  100. package/.agent/skills/intelligent-routing/SKILL.md +9 -23
  101. package/.agent/skills/knowledge-graph/SKILL.md +9 -21
  102. package/.agent/skills/landing-page/SKILL.md +19 -16
  103. package/.agent/skills/lint-and-validate/SKILL.md +9 -59
  104. package/.agent/skills/llm-engineering/SKILL.md +8 -58
  105. package/.agent/skills/local-first/SKILL.md +9 -59
  106. package/.agent/skills/local-first-architecture/SKILL.md +37 -20
  107. package/.agent/skills/lottie-animation/SKILL.md +19 -15
  108. package/.agent/skills/marquee-loop/SKILL.md +19 -16
  109. package/.agent/skills/masked-reveal/SKILL.md +19 -15
  110. package/.agent/skills/mcp-builder/SKILL.md +9 -59
  111. package/.agent/skills/micro-interaction/SKILL.md +19 -16
  112. package/.agent/skills/mobile-design/SKILL.md +8 -58
  113. package/.agent/skills/monorepo-management/SKILL.md +9 -59
  114. package/.agent/skills/morphing-icons/SKILL.md +19 -15
  115. package/.agent/skills/motion-engineering/SKILL.md +22 -15
  116. package/.agent/skills/nextjs-react-expert/SKILL.md +35 -67
  117. package/.agent/skills/nodejs-best-practices/SKILL.md +34 -61
  118. package/.agent/skills/observability/SKILL.md +8 -58
  119. package/.agent/skills/opentelemetry-observability/SKILL.md +31 -5
  120. package/.agent/skills/page-transition-animation/SKILL.md +19 -16
  121. package/.agent/skills/parallel-agents/SKILL.md +9 -59
  122. package/.agent/skills/performance-profiling/SKILL.md +8 -58
  123. package/.agent/skills/plan-writing/SKILL.md +36 -69
  124. package/.agent/skills/platform-engineer/SKILL.md +9 -59
  125. package/.agent/skills/platform-engineering-opentofu/SKILL.md +31 -5
  126. package/.agent/skills/playwright-ai-e2e/SKILL.md +31 -5
  127. package/.agent/skills/playwright-best-practices/SKILL.md +9 -59
  128. package/.agent/skills/polish/SKILL.md +19 -16
  129. package/.agent/skills/powershell-windows/SKILL.md +8 -58
  130. package/.agent/skills/pricing-page/SKILL.md +19 -16
  131. package/.agent/skills/product-aware-heuristics/SKILL.md +27 -8
  132. package/.agent/skills/progressive-blur/SKILL.md +19 -16
  133. package/.agent/skills/project-idioms/SKILL.md +9 -59
  134. package/.agent/skills/property-based-testing/SKILL.md +31 -5
  135. package/.agent/skills/python-patterns/SKILL.md +9 -59
  136. package/.agent/skills/python-pro/SKILL.md +35 -67
  137. package/.agent/skills/quieter/SKILL.md +19 -16
  138. package/.agent/skills/react-doctor/SKILL.md +19 -16
  139. package/.agent/skills/react-specialist/SKILL.md +26 -65
  140. package/.agent/skills/readme-builder/SKILL.md +9 -59
  141. package/.agent/skills/realtime-patterns/SKILL.md +8 -58
  142. package/.agent/skills/red-team-tactics/SKILL.md +9 -59
  143. package/.agent/skills/redesign-skill/SKILL.md +19 -16
  144. package/.agent/skills/review-animations/SKILL.md +21 -25
  145. package/.agent/skills/rust-pro/SKILL.md +30 -61
  146. package/.agent/skills/seo-fundamentals/SKILL.md +9 -59
  147. package/.agent/skills/server-management/SKILL.md +9 -59
  148. package/.agent/skills/shadcn-ui-expert/SKILL.md +9 -59
  149. package/.agent/skills/shape/SKILL.md +19 -16
  150. package/.agent/skills/skill-creator/SKILL.md +36 -86
  151. package/.agent/skills/soft-skill/SKILL.md +19 -16
  152. package/.agent/skills/sounds-on-the-web/SKILL.md +19 -16
  153. package/.agent/skills/sql-pro/SKILL.md +47 -62
  154. package/.agent/skills/supabase-postgres-best-practices/SKILL.md +9 -59
  155. package/.agent/skills/svg-animation/SKILL.md +19 -15
  156. package/.agent/skills/swiftui-expert/SKILL.md +9 -59
  157. package/.agent/skills/swiss-design/SKILL.md +19 -16
  158. package/.agent/skills/system-design-pro/SKILL.md +21 -26
  159. package/.agent/skills/systematic-debugging/SKILL.md +19 -60
  160. package/.agent/skills/systematic-debugging/condition-based-waiting.md +115 -0
  161. package/.agent/skills/systematic-debugging/defense-in-depth.md +122 -0
  162. package/.agent/skills/systematic-debugging/root-cause-tracing.md +169 -0
  163. package/.agent/skills/tailwind-patterns/SKILL.md +8 -58
  164. package/.agent/skills/taste-skill/SKILL.md +19 -16
  165. package/.agent/skills/tdd-workflow/SKILL.md +65 -25
  166. package/.agent/skills/test-result-analyzer/SKILL.md +9 -59
  167. package/.agent/skills/testing-patterns/SKILL.md +8 -58
  168. package/.agent/skills/thermo-nuclear-code-quality-review/SKILL.md +19 -18
  169. package/.agent/skills/thinking-protocol/SKILL.md +33 -2
  170. package/.agent/skills/to-spring-or-not-to-spring/SKILL.md +19 -15
  171. package/.agent/skills/transitions-dev/SKILL.md +19 -15
  172. package/.agent/skills/trend-researcher/SKILL.md +9 -59
  173. package/.agent/skills/typescript-advanced/SKILL.md +21 -64
  174. package/.agent/skills/typeset/SKILL.md +19 -16
  175. package/.agent/skills/ui-reasoning-engine/SKILL.md +27 -8
  176. package/.agent/skills/ui-skill-packs/SKILL.md +26 -16
  177. package/.agent/skills/ui-skills-root/SKILL.md +22 -14
  178. package/.agent/skills/ui-ux-pro-max/SKILL.md +38 -12
  179. package/.agent/skills/ui-ux-researcher/SKILL.md +9 -59
  180. package/.agent/skills/vector-search-pgvector/SKILL.md +31 -5
  181. package/.agent/skills/verification-before-completion/SKILL.md +125 -0
  182. package/.agent/skills/vue-expert/SKILL.md +9 -57
  183. package/.agent/skills/vulnerability-scanner/SKILL.md +8 -58
  184. package/.agent/skills/web-accessibility-auditor/SKILL.md +9 -59
  185. package/.agent/skills/web-design-guidelines/SKILL.md +27 -7
  186. package/.agent/skills/web-quality-audit/SKILL.md +19 -16
  187. package/.agent/skills/webapp-testing/SKILL.md +9 -59
  188. package/.agent/skills/webgpu-performance/SKILL.md +9 -28
  189. package/.agent/skills/whimsy-injector/SKILL.md +9 -57
  190. package/.agent/skills/workflow-optimizer/SKILL.md +9 -57
  191. package/.agent/skills/zero-trust-passkeys/SKILL.md +31 -5
  192. package/.agent/templates/DESIGN.md +680 -160
  193. package/.agent/templates/sdd/implementer-prompt.md +53 -0
  194. package/.agent/templates/sdd/re-review-prompt.md +28 -0
  195. package/.agent/templates/sdd/task-reviewer-prompt.md +59 -0
  196. package/.agent/workflows/acf.md +4 -0
  197. package/.agent/workflows/api-tester.md +5 -0
  198. package/.agent/workflows/audit.md +5 -0
  199. package/.agent/workflows/brainstorm.md +17 -1
  200. package/.agent/workflows/changelog.md +5 -0
  201. package/.agent/workflows/contract.md +19 -0
  202. package/.agent/workflows/create.md +6 -1
  203. package/.agent/workflows/debug.md +5 -0
  204. package/.agent/workflows/deploy.md +6 -1
  205. package/.agent/workflows/enhance.md +6 -1
  206. package/.agent/workflows/fix-ci.md +5 -0
  207. package/.agent/workflows/fix.md +5 -0
  208. package/.agent/workflows/generate.md +5 -0
  209. package/.agent/workflows/marathon.md +6 -1
  210. package/.agent/workflows/migrate.md +5 -0
  211. package/.agent/workflows/minimal.md +4 -0
  212. package/.agent/workflows/orchestrate.md +6 -1
  213. package/.agent/workflows/performance-benchmarker.md +6 -1
  214. package/.agent/workflows/pipeline.md +5 -0
  215. package/.agent/workflows/plan.md +6 -1
  216. package/.agent/workflows/preview.md +6 -1
  217. package/.agent/workflows/refactor.md +6 -0
  218. package/.agent/workflows/review-ai.md +5 -0
  219. package/.agent/workflows/review.md +5 -0
  220. package/.agent/workflows/sdd.md +122 -0
  221. package/.agent/workflows/session.md +5 -0
  222. package/.agent/workflows/status.md +4 -0
  223. package/.agent/workflows/super-prompt.md +4 -0
  224. package/.agent/workflows/swarm.md +6 -1
  225. package/.agent/workflows/test.md +6 -1
  226. package/.agent/workflows/tribunal-backend.md +5 -0
  227. package/.agent/workflows/tribunal-cicd.md +5 -0
  228. package/.agent/workflows/tribunal-database.md +5 -0
  229. package/.agent/workflows/tribunal-full.md +8 -0
  230. package/.agent/workflows/tribunal-mobile.md +5 -0
  231. package/.agent/workflows/tribunal-performance.md +5 -0
  232. package/.agent/workflows/tribunal-speed.md +5 -0
  233. package/.agent/workflows/tribunal-ui.md +7 -0
  234. package/.agents/plugins/marketplace.json +20 -0
  235. package/.claude-plugin/marketplace.json +17 -0
  236. package/.claude-plugin/plugin.json +26 -0
  237. package/.codex-plugin/plugin.json +45 -0
  238. package/.cursor-plugin/plugin.json +26 -0
  239. package/.devin-plugin/plugin.json +14 -0
  240. package/.hermes-plugin/__init__.py +102 -0
  241. package/.hermes-plugin/plugin.yaml +6 -0
  242. package/.kimi-plugin/plugin.json +38 -0
  243. package/.opencode/INSTALL.md +41 -0
  244. package/.opencode/plugins/tribunal.js +101 -0
  245. package/.pi/extensions/tribunal.ts +133 -0
  246. package/CONTRIBUTING.md +1 -1
  247. package/README.md +246 -221
  248. package/bin/wrapper.js +1 -0
  249. package/dist/cli.js +355 -355
  250. package/dist/esm/index.mjs +142 -142
  251. package/dist/index.d.ts +349 -343
  252. package/gemini-extension.json +6 -0
  253. package/hooks/hooks-cursor.json +16 -0
  254. package/hooks/hooks.json +16 -0
  255. package/hooks/session-start.js +77 -0
  256. package/package.json +165 -146
  257. package/scripts/audit_skill_sdo.js +87 -0
  258. package/scripts/modernize_skills.js +254 -0
package/.agent/skills/web-quality-audit/SKILL.md CHANGED
@@ -1,8 +1,8 @@
1
1
  ---
2
2
  name: web-quality-audit
3
3
  description: Web quality auditing skill for Lighthouse-style analysis across Performance, Accessibility, Best Practices, and SEO signals.
4
- version: 3.0.0
5
- last-updated: 2026-07-30
4
+ version: 4.0.0
5
+ last-updated: 2026-09-07
6
6
  skills:
7
7
  - web-design-guidelines
8
8
  - fixing-accessibility
@@ -27,6 +27,12 @@ Before executing web quality audits or Lighthouse-style evaluations, you MUST in
27
27
 
28
28
  Conduct a comprehensive multi-pillar web quality audit covering Performance, Accessibility, Best Practices, and SEO signals.
29
29
 
30
+
31
+ ## Activation Boundaries
32
+
33
+ - **Activate when:** Operating in tasks requiring Web quality auditing skill for Lighthouse-style analysis across Performance, Accessibility, Best Practices, and SEO signals..
34
+ - **DO NOT activate when:** The task falls strictly outside web-quality-audit domain or belongs to a different dedicated specialist.
35
+
30
36
  ---
31
37
 
32
38
  ## 4 Quality Pillars
@@ -54,28 +60,25 @@ Conduct a comprehensive multi-pillar web quality audit covering Performance, Acc
54
60
 
55
61
  ---
56
62
 
57
- ## 🤖 LLM-Specific Traps
63
+ ## 🏛️ Tribunal Verification & Guardrails
58
64
 
59
- 1. **Ignoring Image Dimensions**: Omitting `width` and `height` on images, causing layout shifts (CLS penalties).
60
- 2. **Synchronous Heavy Scripts**: Loading render-blocking third-party scripts in the document `<head>` without `async` or `defer`.
61
-
62
- ---
65
+ **Slash command: `/review` or `/tribunal-full`**
66
+ **Active reviewers: `logic-reviewer` · `security-auditor`**
63
67
 
64
- ## 🏛️ Tribunal Integration (Anti-Hallucination)
65
-
66
- **Active reviewers: `vitals-reviewer` · `accessibility-reviewer` · `seo-specialist`**
68
+ ### ❌ Forbidden AI Tropes
69
+ 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
70
+ 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
71
+ 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
67
72
 
68
73
  ### ✅ Pre-Flight Self-Audit
69
-
70
74
  ```
71
- ✅ Are explicit `width` and `height` aspect ratios declared on all media tags?
72
- ✅ Are script tags in the `<head>` marked with `defer` or `async`?
73
- ✅ Is text contrast WCAG AA compliant across all components?
75
+ ✅ Did I rely ONLY on real, verified tools and methods?
76
+ ✅ Is this solution appropriately scoped to the user's constraints?
77
+ ✅ Did I handle potential failure modes and edge cases?
78
+ ✅ Have I avoided generic boilerplate that doesn't add value?
74
79
  ```
75
80
 
76
81
  ### 🛑 Verification-Before-Completion (VBC) Protocol
77
-
78
82
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
79
-
80
83
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
81
84
  - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
package/.agent/skills/webapp-testing/SKILL.md CHANGED
@@ -2,8 +2,8 @@
2
2
  name: webapp-testing
3
3
  description: Comprehensive Web Application Testing strategy. Test Pyramid, Vitest/Jest for unit logic, React Testing Library for component integrity, MSW (Mock Service Worker) for API layer simulation, and visual regression. Use when setting up testing environments or defining global test strategies across a web stack.
4
4
  tools: Read, Grep, Glob, Bash, Edit, Write
5
- version: 3.0.0
6
- last-updated: 2026-07-30
5
+ version: 4.0.0
6
+ last-updated: 2026-09-07
7
7
  skills:
8
8
  - testing-patterns
9
9
  - playwright-best-practices
@@ -25,6 +25,12 @@ Before building web application test suites or configuring component mocks, you
25
25
  2. MSW Network Interception Rule (Section 40) → Intercept HTTP network calls via MSW (`setupServer`); ban mocking library clients like `axios.get.mockResolvedValue`
26
26
  3. Decouple Business Logic from React (Section 99) → Extract complex calculations into pure testable functions outside React components
27
27
 
28
+
29
+ ## Activation Boundaries
30
+
31
+ - **Activate when:** Operating in tasks requiring Comprehensive Web Application Testing strategy. Test Pyramid, Vitest/Jest for unit logic, React Testing Library for component integrity, MSW (Mock Service Worker) for API layer simulation, and visual regression. Use when setting up testing environments or defining global test strategies across a web stack..
32
+ - **DO NOT activate when:** The task falls strictly outside webapp-testing domain or belongs to a different dedicated specialist.
33
+
28
34
  ## Hallucination Traps (Read First)
29
35
 
30
36
  - ❌ Mocking everything in integration tests -> ✅ Integration tests verify real interactions; mock only external services (APIs, DBs)
@@ -33,8 +39,6 @@ Before building web application test suites or configuring component mocks, you
33
39
 
34
40
  ---
35
41
 
36
- # Webapp Testing — Full Stack Pipeline Mastery
37
-
38
42
  ---
39
43
 
40
44
  ## 1. The Strategy (The Testing Trophy)
@@ -146,69 +150,17 @@ AI coding assistants often fall into specific bad habits when dealing with this
146
150
 
147
151
  ---
148
152
 
149
- **Slash command: `/review` or `/tribunal-full`**
150
- **Active reviewers: `logic-reviewer` · `security-auditor`**
151
-
152
- ### ❌ Forbidden AI Tropes
153
-
154
- 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
155
- 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
156
- 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
157
-
158
- Review these questions before confirming output:
159
-
160
- ```
161
- ✅ Did I rely ONLY on real, verified tools and methods?
162
- ✅ Is this solution appropriately scoped to the user's constraints?
163
- ✅ Did I handle potential failure modes and edge cases?
164
- ✅ Have I avoided generic boilerplate that doesn't add value?
165
- ```
166
-
167
- ### 🛑 Verification-Before-Completion (VBC) Protocol
168
-
169
- **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
170
-
171
- - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
172
- - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
173
-
174
- ## Pre-Flight Checklist
175
-
176
- - [ ] Have I reviewed the user's specific constraints and requests?
177
- - [ ] Have I checked the environment for relevant existing implementations?
178
-
179
- ## VBC Protocol (Verification-Before-Completion)
180
-
181
- You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
182
-
183
- ---
184
-
185
- ## 🤖 LLM-Specific Traps
186
-
187
- AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
188
-
189
- 1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
190
- 2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
191
- 3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
192
- 4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
193
- 5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
194
-
195
- ---
196
-
197
- ## 🏛️ Tribunal Integration (Anti-Hallucination)
153
+ ## 🏛️ Tribunal Verification & Guardrails
198
154
 
199
155
  **Slash command: `/review` or `/tribunal-full`**
200
156
  **Active reviewers: `logic-reviewer` · `security-auditor`**
201
157
 
202
158
  ### ❌ Forbidden AI Tropes
203
-
204
159
  1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
205
160
  2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
206
161
  3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
207
162
 
208
163
  ### ✅ Pre-Flight Self-Audit
209
-
210
- Review these questions before confirming output:
211
-
212
164
  ```
213
165
  ✅ Did I rely ONLY on real, verified tools and methods?
214
166
  ✅ Is this solution appropriately scoped to the user's constraints?
@@ -217,8 +169,6 @@ Review these questions before confirming output:
217
169
  ```
218
170
 
219
171
  ### 🛑 Verification-Before-Completion (VBC) Protocol
220
-
221
172
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
222
-
223
173
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
224
174
  - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
package/.agent/skills/webgpu-performance/SKILL.md CHANGED
@@ -2,8 +2,8 @@
2
2
  name: webgpu-performance
3
3
  description: High-performance browser graphics and compute mastery. Transitioning from WebGL to WebGPU API, WGSL compute shaders, explicit GPU memory management, and browser-side tensor calculations.
4
4
  tools: Read, Grep, Glob, Bash, Edit, Write
5
- version: 3.0.0
6
- last-updated: 2026-07-30
5
+ version: 4.0.0
6
+ last-updated: 2026-09-07
7
7
  skills:
8
8
  - 60fps-animation
9
9
  - fixing-motion-performance
@@ -27,6 +27,12 @@ Before implementing WebGPU shaders or pipelines, you MUST inspect:
27
27
 
28
28
  You are an expert in writing low-level, high-performance browser graphics and compute pipelines using WebGPU and WGSL (WebGPU Shading Language). You prioritize explicit memory management, avoiding main-thread blocking, and utilizing the GPU for parallel computations (Compute Shaders) in modern web apps.
29
29
 
30
+
31
+ ## Activation Boundaries
32
+
33
+ - **Activate when:** Operating in tasks requiring High-performance browser graphics and compute mastery. Transitioning from WebGL to WebGPU API, WGSL compute shaders, explicit GPU memory management, and browser-side tensor calculations..
34
+ - **DO NOT activate when:** The task falls strictly outside webgpu-performance domain or belongs to a different dedicated specialist.
35
+
30
36
  ## 1. Core Principles
31
37
 
32
38
  - **Explicit > Implicit:** Unlike WebGL, WebGPU doesn't hide state. You must explicitly configure Pipelines, BindGroups, and CommandEncoders.
@@ -93,42 +99,19 @@ Before submitting code, ensure:
93
99
  2. WGSL workgroup sizes align with the dispatch sizes dynamically.
94
100
  3. GPUBuffers used for compute have `GPUBufferUsage.STORAGE` flags.
95
101
 
96
- ### 🛑 Verification-Before-Completion (VBC) Protocol
97
-
98
- **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
99
-
100
- - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
101
- - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
102
-
103
102
  ---
104
103
 
105
- ## 🤖 LLM-Specific Traps
106
-
107
- AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
108
-
109
- 1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
110
- 2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
111
- 3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
112
- 4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
113
- 5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
114
-
115
- ---
116
-
117
- ## 🏛️ Tribunal Integration (Anti-Hallucination)
104
+ ## 🏛️ Tribunal Verification & Guardrails
118
105
 
119
106
  **Slash command: `/review` or `/tribunal-full`**
120
107
  **Active reviewers: `logic-reviewer` · `security-auditor`**
121
108
 
122
109
  ### ❌ Forbidden AI Tropes
123
-
124
110
  1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
125
111
  2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
126
112
  3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
127
113
 
128
114
  ### ✅ Pre-Flight Self-Audit
129
-
130
- Review these questions before confirming output:
131
-
132
115
  ```
133
116
  ✅ Did I rely ONLY on real, verified tools and methods?
134
117
  ✅ Is this solution appropriately scoped to the user's constraints?
@@ -137,8 +120,6 @@ Review these questions before confirming output:
137
120
  ```
138
121
 
139
122
  ### 🛑 Verification-Before-Completion (VBC) Protocol
140
-
141
123
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
142
-
143
124
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
144
125
  - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
package/.agent/skills/whimsy-injector/SKILL.md CHANGED
@@ -2,8 +2,8 @@
2
2
  name: whimsy-injector
3
3
  description: Micro-delight generator for frontend interfaces. Suggests and implements subtle animations, playful transitions, and interaction polish across any frontend stack.
4
4
  tools: Read, Grep, Glob, Bash, Edit, Write
5
- version: 3.0.0
6
- last-updated: 2026-07-30
5
+ version: 4.0.0
6
+ last-updated: 2026-09-07
7
7
  skills:
8
8
  - delight
9
9
  - micro-interaction
@@ -28,6 +28,12 @@ Before injecting micro-animations or interaction polish, you MUST inspect:
28
28
 
29
29
  You are a specialized agent for injecting "micro-delights" into user interfaces — small, tasteful animations and interaction polish that make an app feel alive and premium. You work across all frontend stacks: vanilla CSS, React, Vue, Svelte, Angular, and static HTML.
30
30
 
31
+
32
+ ## Activation Boundaries
33
+
34
+ - **Activate when:** Operating in tasks requiring Micro-delight generator for frontend interfaces. Suggests and implements subtle animations, playful transitions, and interaction polish across any frontend stack..
35
+ - **DO NOT activate when:** The task falls strictly outside whimsy-injector domain or belongs to a different dedicated specialist.
36
+
31
37
  ## When to Activate
32
38
 
33
39
  - During `/enhance` or `/ui-ux-pro-max` workflows as a polish pass.
@@ -324,69 +330,17 @@ AI coding assistants often fall into specific bad habits when dealing with this
324
330
 
325
331
  ---
326
332
 
327
- **Slash command: `/review` or `/tribunal-full`**
328
- **Active reviewers: `logic-reviewer` · `security-auditor`**
329
-
330
- ### ❌ Forbidden AI Tropes
331
-
332
- 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
333
- 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
334
- 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
335
-
336
- Review these questions before confirming output:
337
-
338
- ```
339
- ✅ Did I rely ONLY on real, verified tools and methods?
340
- ✅ Is this solution appropriately scoped to the user's constraints?
341
- ✅ Did I handle potential failure modes and edge cases?
342
- ✅ Have I avoided generic boilerplate that doesn't add value?
343
- ```
344
-
345
- ### 🛑 Verification-Before-Completion (VBC) Protocol
346
-
347
- **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
348
-
349
- - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
350
- - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
351
-
352
- ## Pre-Flight Checklist
353
-
354
- - [ ] Have I reviewed the user's specific constraints and requests?
355
- - [ ] Have I checked the environment for relevant existing implementations?
356
-
357
- ## VBC Protocol (Verification-Before-Completion)
358
-
359
- You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
360
-
361
- ---
362
-
363
- ## 🤖 LLM-Specific Traps
364
-
365
- AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
366
-
367
- 1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
368
- 2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
369
- 3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
370
- 4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
371
- 5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
372
-
373
- ---
374
-
375
- ## 🏛️ Tribunal Integration (Anti-Hallucination)
333
+ ## 🏛️ Tribunal Verification & Guardrails
376
334
 
377
335
  **Slash command: `/review` or `/tribunal-full`**
378
336
  **Active reviewers: `logic-reviewer` · `security-auditor`**
379
337
 
380
338
  ### ❌ Forbidden AI Tropes
381
-
382
339
  1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
383
340
  2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
384
341
  3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
385
342
 
386
343
  ### ✅ Pre-Flight Self-Audit
387
-
388
- Review these questions before confirming output:
389
-
390
344
  ```
391
345
  ✅ Did I rely ONLY on real, verified tools and methods?
392
346
  ✅ Is this solution appropriately scoped to the user's constraints?
@@ -395,8 +349,6 @@ Review these questions before confirming output:
395
349
  ```
396
350
 
397
351
  ### 🛑 Verification-Before-Completion (VBC) Protocol
398
-
399
352
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
400
-
401
353
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
402
354
  - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
package/.agent/skills/workflow-optimizer/SKILL.md CHANGED
@@ -1,8 +1,8 @@
1
1
  ---
2
2
  name: workflow-optimizer
3
3
  description: Analyzes agent tool-calling patterns and task execution efficiency to suggest process improvements.
4
- version: 3.0.0
5
- last-updated: 2026-07-30
4
+ version: 4.0.0
5
+ last-updated: 2026-09-07
6
6
  skills:
7
7
  - parallel-agents
8
8
  - plan-writing
@@ -27,6 +27,12 @@ Before analyzing tool-calling logs or task efficiency, you MUST inspect:
27
27
 
28
28
  You are a specialized agent for analyzing and optimizing the efficiency of AI agent workflows, task execution loops, and tool-calling patterns. You act as a "meta-debugger" — debugging the _process_, not the _code_.
29
29
 
30
+
31
+ ## Activation Boundaries
32
+
33
+ - **Activate when:** Operating in tasks requiring Analyzes agent tool-calling patterns and task execution efficiency to suggest process improvements..
34
+ - **DO NOT activate when:** The task falls strictly outside workflow-optimizer domain or belongs to a different dedicated specialist.
35
+
30
36
  ## When to Activate
31
37
 
32
38
  - When a task takes significantly more tool calls than expected.
@@ -206,69 +212,17 @@ AI coding assistants often fall into specific bad habits when dealing with this
206
212
 
207
213
  ---
208
214
 
209
- **Slash command: `/review` or `/tribunal-full`**
210
- **Active reviewers: `logic-reviewer` · `security-auditor`**
211
-
212
- ### ❌ Forbidden AI Tropes
213
-
214
- 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
215
- 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
216
- 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
217
-
218
- Review these questions before confirming output:
219
-
220
- ```
221
- ✅ Did I rely ONLY on real, verified tools and methods?
222
- ✅ Is this solution appropriately scoped to the user's constraints?
223
- ✅ Did I handle potential failure modes and edge cases?
224
- ✅ Have I avoided generic boilerplate that doesn't add value?
225
- ```
226
-
227
- ### 🛑 Verification-Before-Completion (VBC) Protocol
228
-
229
- **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
230
-
231
- - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
232
- - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
233
-
234
- ## Pre-Flight Checklist
235
-
236
- - [ ] Have I reviewed the user's specific constraints and requests?
237
- - [ ] Have I checked the environment for relevant existing implementations?
238
-
239
- ## VBC Protocol (Verification-Before-Completion)
240
-
241
- You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
242
-
243
- ---
244
-
245
- ## 🤖 LLM-Specific Traps
246
-
247
- AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
248
-
249
- 1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
250
- 2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
251
- 3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
252
- 4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
253
- 5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
254
-
255
- ---
256
-
257
- ## 🏛️ Tribunal Integration (Anti-Hallucination)
215
+ ## 🏛️ Tribunal Verification & Guardrails
258
216
 
259
217
  **Slash command: `/review` or `/tribunal-full`**
260
218
  **Active reviewers: `logic-reviewer` · `security-auditor`**
261
219
 
262
220
  ### ❌ Forbidden AI Tropes
263
-
264
221
  1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
265
222
  2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
266
223
  3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
267
224
 
268
225
  ### ✅ Pre-Flight Self-Audit
269
-
270
- Review these questions before confirming output:
271
-
272
226
  ```
273
227
  ✅ Did I rely ONLY on real, verified tools and methods?
274
228
  ✅ Is this solution appropriately scoped to the user's constraints?
@@ -277,8 +231,6 @@ Review these questions before confirming output:
277
231
  ```
278
232
 
279
233
  ### 🛑 Verification-Before-Completion (VBC) Protocol
280
-
281
234
  **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
282
-
283
235
  - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
284
236
  - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.
package/.agent/skills/zero-trust-passkeys/SKILL.md CHANGED
@@ -2,8 +2,8 @@
2
2
  name: zero-trust-passkeys
3
3
  description: Modern passwordless authentication using WebAuthn, FIDO2 biometric passkeys, SimpleWebAuthn v13+, Conditional UI (Passkey Autofill), and Zero-Trust security.
4
4
  tools: Read, Grep, Glob, Edit, Write
5
- version: 3.0.0
6
- last-updated: 2026-08-05
5
+ version: 4.0.0
6
+ last-updated: 2026-09-07
7
7
  script: .agent/scripts/security_scan.js
8
8
  scripts-binding:
9
9
  - .agent/scripts/security_scan.js
@@ -23,6 +23,12 @@ Before implementing auth flows:
23
23
  2. Feature Detection & Abort Signals → Check `isConditionalMediationAvailable()` and manage cancellation via `AbortController`
24
24
  3. Discoverable Credentials → Ensure `userVerification` and resident key options are enabled on registration
25
25
 
26
+
27
+ ## Activation Boundaries
28
+
29
+ - **Activate when:** Operating in tasks requiring Modern passwordless authentication using WebAuthn, FIDO2 biometric passkeys, SimpleWebAuthn v13+, Conditional UI (Passkey Autofill), and Zero-Trust security..
30
+ - **DO NOT activate when:** The task falls strictly outside zero-trust-passkeys domain or belongs to a different dedicated specialist.
31
+
26
32
  ## Client-Side Passkey Autofill Pattern (SimpleWebAuthn v13 Browser)
27
33
 
28
34
  ```typescript
@@ -84,7 +90,27 @@ export async function verifyPasskeyAuth(
84
90
  }
85
91
  ```
86
92
 
87
- ## 🛑 Verification-Before-Completion (VBC) Protocol
93
+ ---
94
+
95
+ ## 🏛️ Tribunal Verification & Guardrails
96
+
97
+ **Slash command: `/review` or `/tribunal-full`**
98
+ **Active reviewers: `logic-reviewer` · `security-auditor`**
99
+
100
+ ### ❌ Forbidden AI Tropes
101
+ 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
102
+ 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
103
+ 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
104
+
105
+ ### ✅ Pre-Flight Self-Audit
106
+ ```
107
+ ✅ Did I rely ONLY on real, verified tools and methods?
108
+ ✅ Is this solution appropriately scoped to the user's constraints?
109
+ ✅ Did I handle potential failure modes and edge cases?
110
+ ✅ Have I avoided generic boilerplate that doesn't add value?
111
+ ```
88
112
 
89
- - Verify input tags have `autocomplete="username webauthn"`.
90
- - Test passkey autofill flow with `AbortController` cancellation.
113
+ ### 🛑 Verification-Before-Completion (VBC) Protocol
114
+ **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
115
+ - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
116
+ - ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing tests, compile success, or equivalent proof) that your output works as intended.