tribunal-kit 8.0.0 → 9.0.0

package/.agent/skills/local-first-architecture/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: local-first-architecture
 description: Local-first architecture guidance for instant-feeling web & desktop apps, optimistic updates, IndexedDB sync engines, offline operation, and Conflict-free Replicated Data Types (CRDTs).
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - react-specialist
   - nextjs-react-expert
@@ -21,11 +21,31 @@ scripts-binding:
 
 Before implementing local-first architecture or offline sync engines, you MUST inspect:
 
-1. Zero Network Latency Rule (Section 24) → Execute reads and writes against local storage (IndexedDB/SQLite) first before queueing outbound sync
-2. Outbound Mutation Queue (Section 28) → Persist mutation logs in IndexedDB with client-generated UUID `idempotencyKey` fields
-3. Conflict Resolution Strategy (Section 33) → Use Last-Write-Wins (LWW) for simple fields or CRDTs (Yjs/Automerge) for collaborative state
+1. Zero Network Latency Rule → Execute reads and writes against local storage (OPFS / IndexedDB / WASM SQLite) first before queueing outbound sync
+2. Outbound Mutation Queue & Idempotency → Persist mutation logs locally with client-generated UUID `idempotencyKey` fields to prevent duplicate server writes
+3. Conflict Resolution Strategy → Define explicit conflict handling (Last-Write-Wins with Lamport timestamps or CRDTs via Yjs/Automerge)
+4. OPFS (Origin Private File System) → In modern web browsers, use OPFS for high-speed disk I/O instead of raw legacy IndexedDB for database files
 
-Architect software that reads and writes to local storage (SQLite/IndexedDB) first, ensuring instant zero-latency UI mutations with background server synchronization.
+## Activation Boundaries
+
+- **Activate when:** Building instant-feeling offline-capable web or desktop applications, collaborative editors, sync engines, and optimistic data stores.
+- **DO NOT activate when:** Building traditional request-response server-rendered apps where clients hold zero persistent state.
+
+## 2026 Local-First Architecture & Sync Invariants
+
+1. **OPFS (Origin Private File System) for WASM SQLite**:
+   Always prefer OPFS sync access handles for SQLite in web workers. It delivers native near-SSD read/write speeds that dwarf standard IndexedDB overhead.
+2. **Idempotency Key Protocol**:
+   Every local mutation must generate an `idempotencyKey` (UUID v7). The server stores processed keys for 24h, discarding replays during network flushes.
+3. **Delta Sync over Full Sync**:
+   Never download entire tables. Use incremental sync cursors (`last_synced_version_id`) and stream compact binary deltas via WebSockets or SSE.
+
+## Hallucination Traps (Read First)
+
+- ❌ Blocking UI render on a remote fetch in a local-first app → ✅ Render local cache immediately; reconcile in background
+- ❌ Mutating server state without an idempotency key → ✅ Flushed network retries will duplicate records
+- ❌ Putting heavy SQLite WASM operations on the browser main thread → ✅ Always run local DB queries inside a Dedicated Web Worker
+- ❌ Storing unindexed large blobs inside IndexedDB → ✅ Store blobs in OPFS; keep metadata in local SQLite/IndexedDB
 
 ---
 
@@ -53,28 +73,25 @@ Architect software that reads and writes to local storage (SQLite/IndexedDB) fir
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🏛️ Tribunal Verification & Guardrails
 
-1. **Waiting for Server API Response Before UI Render**: Blocking the user interface with a spinner while waiting for a POST request response.
-2. **Volatile Local State**: Keeping mutation queues in non-persistent JS memory, losing unsaved user edits on page refresh.
+**Slash command: `/review` or `/tribunal-full`**
+**Active reviewers: `logic-reviewer` · `security-auditor`**
 
----
-
-## 🏛️ Tribunal Integration (Anti-Hallucination)
-
-**Active reviewers: `logic-reviewer` · `resilience-reviewer` · `architecture-reviewer`**
+### ❌ Forbidden AI Tropes
+1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
+2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
+3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
 ```
-✅ Are user actions reflected instantly in local UI state?
-✅ Is the mutation queue persisted in IndexedDB/SQLite for offline durability?
-✅ Are all server API mutations idempotent with client-generated UUID keys?
+✅ Did I rely ONLY on real, verified tools and methods?
+✅ Is this solution appropriately scoped to the user's constraints?
+✅ Did I handle potential failure modes and edge cases?
+✅ Have I avoided generic boilerplate that doesn't add value?
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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/lottie-animation/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: lottie-animation
 description: Lottie and dotLottie integration, playback control, hover/click triggers, runtime theming, and performance optimization for React/Vue/Web.
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - 60fps-animation
   - motion-engineering
@@ -27,6 +27,12 @@ Before integrating Lottie animations, you MUST inspect:
 
 Integrate lightweight, vector-based Lottie animations with interactive trigger controls and runtime color theming.
 
+
+## Activation Boundaries
+
+- **Activate when:** Operating in tasks requiring Lottie and dotLottie integration, playback control, hover/click triggers, runtime theming, and performance optimization for React/Vue/Web..
+- **DO NOT activate when:** The task falls strictly outside lottie-animation domain or belongs to a different dedicated specialist.
+
 ---
 
 ## 3 Lottie Integration Recipes
@@ -69,27 +75,25 @@ Pass custom CSS variable overrides to dotLottie players to dynamically recolor v
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🏛️ Tribunal Verification & Guardrails
 
-1. **Embedding Massive Uncompressed Lottie JSONs**: Importing 2MB JSON files inline into JavaScript bundles.
-2. **Autoplay Loops Everywhere**: Setting `autoplay loop` on 10 offscreen icons, causing massive CPU idle usage.
-
----
+**Slash command: `/review` or `/tribunal-full`**
+**Active reviewers: `logic-reviewer` · `security-auditor`**
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
-
-**Active reviewers: `frontend-reviewer` · `performance-optimizer`**
+### ❌ Forbidden AI Tropes
+1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
+2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
+3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
 ```
-✅ Are dotLottie (`.lottie`) binary assets preferred over raw uncompressed `.json`?
-✅ Is playback triggered intentionally via hover, click, or visible viewport intersection?
+✅ Did I rely ONLY on real, verified tools and methods?
+✅ Is this solution appropriately scoped to the user's constraints?
+✅ Did I handle potential failure modes and edge cases?
+✅ Have I avoided generic boilerplate that doesn't add value?
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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/marquee-loop/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: marquee-loop
 description: Hardware-accelerated, seamless, continuous looping marquees for logo rows, testimonials, and announcements with pause-on-hover accessibility.
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - 60fps-animation
   - company-logos
@@ -27,6 +27,12 @@ Before implementing marquee loop tracks, you MUST inspect:
 
 Build seamless 60fps infinite marquee tracks that pause on hover and respect accessibility settings.
 
+
+## Activation Boundaries
+
+- **Activate when:** Operating in tasks requiring Hardware-accelerated, seamless, continuous looping marquees for logo rows, testimonials, and announcements with pause-on-hover accessibility..
+- **DO NOT activate when:** The task falls strictly outside marquee-loop domain or belongs to a different dedicated specialist.
+
 ---
 
 ## The Seamless CSS Marquee Recipe
@@ -79,28 +85,25 @@ Build seamless 60fps infinite marquee tracks that pause on hover and respect acc
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🏛️ Tribunal Verification & Guardrails
 
-1. **Missing Duplicate Duplicate Track**: Animating a single track without duplicating content, causing a visible blank jump when the animation loops.
-2. **Missing `animation-play-state: paused` on Hover**: Blocking users from clicking or reading marquee content by refusing to pause on hover/focus.
-
----
+**Slash command: `/review` or `/tribunal-full`**
+**Active reviewers: `logic-reviewer` · `security-auditor`**
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
-
-**Active reviewers: `frontend-reviewer` · `performance-optimizer`**
+### ❌ Forbidden AI Tropes
+1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
+2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
+3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
 ```
-✅ Is the duplicate item track present with `aria-hidden="true"`?
-✅ Is `animation-play-state: paused` active on container hover/focus?
-✅ Does marquee collapse into static grid when `prefers-reduced-motion` is enabled?
+✅ Did I rely ONLY on real, verified tools and methods?
+✅ Is this solution appropriately scoped to the user's constraints?
+✅ Did I handle potential failure modes and edge cases?
+✅ Have I avoided generic boilerplate that doesn't add value?
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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/masked-reveal/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: masked-reveal
 description: CSS clip-path, SVG masking, and layered entrance reveals for images, hero banners, and text blocks.
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - 60fps-animation
   - framer-motion-expert
@@ -27,6 +27,12 @@ Before implementing clip-path or SVG mask reveals, you MUST inspect:
 
 Create editorial, high-end visual reveals using CSS `clip-path` and SVG masks.
 
+
+## Activation Boundaries
+
+- **Activate when:** Operating in tasks requiring CSS clip-path, SVG masking, and layered entrance reveals for images, hero banners, and text blocks..
+- **DO NOT activate when:** The task falls strictly outside masked-reveal domain or belongs to a different dedicated specialist.
+
 ---
 
 ## 3 Masked Reveal Patterns
@@ -84,27 +90,25 @@ Wrap text lines inside overflow-hidden wrappers and animate inner text lines upw
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🏛️ Tribunal Verification & Guardrails
 
-1. **Missing `overflow: hidden` on Line Wrappers**: Text showing outside container boundaries during line wipe animations.
-2. **Heavy GPU Mask Layers**: Applying multiple full-screen SVG masks simultaneously on lower-end mobile devices.
-
----
+**Slash command: `/review` or `/tribunal-full`**
+**Active reviewers: `logic-reviewer` · `security-auditor`**
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
-
-**Active reviewers: `motion-reviewer` · `frontend-reviewer`**
+### ❌ Forbidden AI Tropes
+1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
+2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
+3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
 ```
-✅ Are text line wipes enclosed inside `overflow: hidden` wrappers?
-✅ Is `clip-path: inset()` used for linear curtain wipes for maximum performance?
+✅ Did I rely ONLY on real, verified tools and methods?
+✅ Is this solution appropriately scoped to the user's constraints?
+✅ Did I handle potential failure modes and edge cases?
+✅ Have I avoided generic boilerplate that doesn't add value?
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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/mcp-builder/SKILL.md

@@ -2,8 +2,8 @@
 name: mcp-builder
 description: Model Context Protocol (MCP) server integration mastery. Building custom MCP servers, standardizing tool exposes, managing standardized communication between large language models and localized datasets, securing boundary contexts, and architecting resource schemas. Use when modifying, extending, or building custom toolsets for AI platforms relying on the MCP standard.
 tools: Read, Grep, Glob, Bash, Edit, Write
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - backend-security-expert
   - nodejs-best-practices
@@ -25,6 +25,12 @@ Before architecting MCP tools or resource handlers, you MUST inspect:
 2. Resource vs Tool Distinction (Section 68) → Use Resources for static read-only data (`file:///`, `db://schema`), Tools strictly for dynamic parameterized actions
 3. Output Payload Truncation Safeguard (Section 94) → Forcibly truncate large tool response outputs before returning payload to protect LLM context windows
 
+
+## Activation Boundaries
+
+- **Activate when:** Operating in tasks requiring Model Context Protocol (MCP) server integration mastery. Building custom MCP servers, standardizing tool exposes, managing standardized communication between large language models and localized datasets, securing boundary contexts, and architecting resource schemas. Use when modifying, extending, or building custom toolsets for AI platforms relying on the MCP standard..
+- **DO NOT activate when:** The task falls strictly outside mcp-builder domain or belongs to a different dedicated specialist.
+
 ## Hallucination Traps (Read First)
 
 - ❌ Exposing tools without input validation schemas -> ✅ Every MCP tool MUST have JSON Schema for parameters; the protocol requires it
@@ -33,8 +39,6 @@ Before architecting MCP tools or resource handlers, you MUST inspect:
 
 ---
 
-# MCP Builder — Context Protocol Mastery
-
 ---
 
 ## 1. The Anatomy of an MCP Server
@@ -124,69 +128,17 @@ AI coding assistants often fall into specific bad habits when dealing with this
 
 ---
 
-**Slash command: `/review` or `/tribunal-full`**
-**Active reviewers: `logic-reviewer` · `security-auditor`**
-
-### ❌ Forbidden AI Tropes
-
-1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
-2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
-3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
-
-Review these questions before confirming output:
-
-```
-✅ Did I rely ONLY on real, verified tools and methods?
-✅ Is this solution appropriately scoped to the user's constraints?
-✅ Did I handle potential failure modes and edge cases?
-✅ Have I avoided generic boilerplate that doesn't add value?
-```
-
-### 🛑 Verification-Before-Completion (VBC) Protocol
-
-**CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
-- ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
-- ✅ **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.
-
-## Pre-Flight Checklist
-
-- [ ] Have I reviewed the user's specific constraints and requests?
-- [ ] Have I checked the environment for relevant existing implementations?
-
-## VBC Protocol (Verification-Before-Completion)
-
-You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
-
----
-
-## 🤖 LLM-Specific Traps
-
-AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
-
-1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
-2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
-3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
-4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
-5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
-
----
-
-## 🏛️ Tribunal Integration (Anti-Hallucination)
+## 🏛️ Tribunal Verification & Guardrails
 
 **Slash command: `/review` or `/tribunal-full`**
 **Active reviewers: `logic-reviewer` · `security-auditor`**
 
 ### ❌ Forbidden AI Tropes
-
 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
-Review these questions before confirming output:
-
 ```
 ✅ Did I rely ONLY on real, verified tools and methods?
 ✅ Is this solution appropriately scoped to the user's constraints?
@@ -195,8 +147,6 @@ Review these questions before confirming output:
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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/micro-interaction/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: micro-interaction
 description: Detailed UI motion guidance for hover and press feedback, toggles, checkboxes, toasts, drawers, modals, list transitions, and shared-element interactions.
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - delight
   - better-ui
@@ -27,6 +27,12 @@ Before engineering component micro-interactions, you MUST inspect:
 
 Craft responsive, physically grounded micro-interactions for everyday UI controls.
 
+
+## Activation Boundaries
+
+- **Activate when:** Operating in tasks requiring Detailed UI motion guidance for hover and press feedback, toggles, checkboxes, toasts, drawers, modals, list transitions, and shared-element interactions..
+- **DO NOT activate when:** The task falls strictly outside micro-interaction domain or belongs to a different dedicated specialist.
+
 ---
 
 ## 4 Micro-Interaction Recipes
@@ -97,28 +103,25 @@ Popovers scale out from their trigger source using CSS variables:
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🏛️ Tribunal Verification & Guardrails
 
-1. **Center Scaling Popovers**: Popovers springing out from the screen center instead of scaling from their trigger origin.
-2. **Slow Micro-Interactions**: Setting button press or toggle transitions to 400ms, making the UI feel sluggish.
-
----
+**Slash command: `/review` or `/tribunal-full`**
+**Active reviewers: `logic-reviewer` · `security-auditor`**
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
-
-**Active reviewers: `frontend-reviewer` · `motion-reviewer`**
+### ❌ Forbidden AI Tropes
+1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
+2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
+3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
 ```
-✅ Are interactive component feedback durations under 180ms?
-✅ Are popover/dropdown transform origins explicitly tied to their trigger?
-✅ Is checkmark SVG stroke-dasharray used for crisp checkbox check animation?
+✅ Did I rely ONLY on real, verified tools and methods?
+✅ Is this solution appropriately scoped to the user's constraints?
+✅ Did I handle potential failure modes and edge cases?
+✅ Have I avoided generic boilerplate that doesn't add value?
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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/mobile-design/SKILL.md

@@ -2,8 +2,8 @@
 name: mobile-design
 description: Mobile-first design for iOS, Android, Foldables, React Native, Flutter. Touch interaction, haptics, 120Hz performance, on-device AI, spatial UI, Reanimated 3. Use when building mobile UI, animations, or cross-platform apps.
 tools: Read, Grep, Glob, Bash, Edit, Write
-version: 3.0.0
-last-updated: 2026-07-30
+version: 4.0.0
+last-updated: 2026-09-07
 skills:
   - building-native-ui
   - adapt
@@ -25,7 +25,11 @@ Before designing mobile interfaces or writing React Native code, you MUST inspec
 2. List Optimization (Section 32) → Use `@shopify/flash-list` with required `estimatedItemSize` instead of `ScrollView` or `FlatList`
 3. Safe Area Insets (Section 100) → Always wrap content with `useSafeAreaInsets()` to accommodate notch, dynamic island, and home indicators
 
-# Mobile Design — Dense Reference
+
+## Activation Boundaries
+
+- **Activate when:** Operating in tasks requiring Mobile-first design for iOS, Android, Foldables, React Native, Flutter. Touch interaction, haptics, 120Hz performance, on-device AI, spatial UI, Reanimated 3. Use when building mobile UI, animations, or cross-platform apps..
+- **DO NOT activate when:** The task falls strictly outside mobile-design domain or belongs to a different dedicated specialist.
 
 ## Hallucination Traps (Read First)
 
@@ -262,69 +266,17 @@ AI coding assistants often fall into specific bad habits when dealing with this
 
 ---
 
-**Slash command: `/review` or `/tribunal-full`**
-**Active reviewers: `logic-reviewer` · `security-auditor`**
-
-### ❌ Forbidden AI Tropes
-
-1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
-2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
-3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
-
-Review these questions before confirming output:
-
-```
-✅ Did I rely ONLY on real, verified tools and methods?
-✅ Is this solution appropriately scoped to the user's constraints?
-✅ Did I handle potential failure modes and edge cases?
-✅ Have I avoided generic boilerplate that doesn't add value?
-```
-
-### 🛑 Verification-Before-Completion (VBC) Protocol
-
-**CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
-- ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
-- ✅ **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.
-
-## Pre-Flight Checklist
-
-- [ ] Have I reviewed the user's specific constraints and requests?
-- [ ] Have I checked the environment for relevant existing implementations?
-
-## VBC Protocol (Verification-Before-Completion)
-
-You MUST verify existing code signatures and variables before attempting to modify or call them. No hallucination is permitted.
-
----
-
-## 🤖 LLM-Specific Traps
-
-AI coding assistants often fall into specific bad habits when dealing with this domain. These are strictly forbidden:
-
-1. **Over-engineering:** Proposing complex abstractions or distributed systems when a simpler approach suffices.
-2. **Hallucinated Libraries/Methods:** Using non-existent methods or packages. Always `// VERIFY` or check `package.json` / `requirements.txt`.
-3. **Skipping Edge Cases:** Writing the "happy path" and ignoring error handling, timeouts, or data validation.
-4. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
-5. **Silent Degradation:** Catching and suppressing errors without logging or re-raising.
-
----
-
-## 🏛️ Tribunal Integration (Anti-Hallucination)
+## 🏛️ Tribunal Verification & Guardrails
 
 **Slash command: `/review` or `/tribunal-full`**
 **Active reviewers: `logic-reviewer` · `security-auditor`**
 
 ### ❌ Forbidden AI Tropes
-
 1. **Blind Assumptions:** Never make an assumption without documenting it clearly with `// VERIFY: [reason]`.
 2. **Silent Degradation:** Catching and suppressing errors without logging or handling.
 3. **Context Amnesia:** Forgetting the user's constraints and offering generic advice instead of tailored solutions.
 
 ### ✅ Pre-Flight Self-Audit
-
-Review these questions before confirming output:
-
 ```
 ✅ Did I rely ONLY on real, verified tools and methods?
 ✅ Is this solution appropriately scoped to the user's constraints?
@@ -333,8 +285,6 @@ Review these questions before confirming output:
 ```
 
 ### 🛑 Verification-Before-Completion (VBC) Protocol
-
 **CRITICAL:** You must follow a strict "evidence-based closeout" state machine.
-
 - ❌ **Forbidden:** Declaring a task complete because the output "looks correct."
 - ✅ **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.