tribunal-kit 8.0.0 → 9.0.1

package/.agent/skills/accessible-animation/SKILL.md

@@ -1,99 +1,266 @@
 ---
 name: accessible-animation
-description: Tiered reduced-motion patterns for CSS, GSAP, Framer Motion, Lenis, and View Transitions API to ensure full WCAG 2.2 accessibility compliance for motion.
-version: 3.0.0
-last-updated: 2026-07-30
+description: Use when implementing WCAG 2.2 AA compliant UI motion, tiered reduced-motion patterns, vestibular trigger suppression, and accessible transitions across CSS, React, Framer Motion, and GSAP.
+version: 5.0.0
+last-updated: 2026-09-13
 skills:
   - 60fps-animation
   - motion-engineering
   - framer-motion-expert
+  - review-animations
 tools: Read, Grep, Glob, Bash, Edit, Write
 scripts-binding:
   - .agent/scripts/lint_runner.js
   - .agent/scripts/verify_all.js
 ---
 
-# Accessible Animation — Tiered Reduced-Motion Patterns
+# Accessible Animation — Tiered Reduced-Motion Architecture & WCAG 2.2 AA Compliance
+
+## Mandatory Pre-Flight Context Inspection
+
+Before generating, refactoring, or reviewing code in the `accessible-animation` domain, inspect these 5 critical parameters:
+
+1. **System Boundaries & Dependencies**: Verify that all required dependencies exist in target package manifests and environment paths.
+2. **Runtime Context & Platform Invariants**: Confirm target platform constraints (Node.js, Browser, Mobile OS, Edge runtime) before applying APIs.
+3. **Execution Guardrails**: Identify potential side-effects, state mutations, and unhandled asynchronous exceptions.
+4. **Validation & Type Contracts**: Validate input data schemas and strict type constraints across all module interfaces.
+5. **Observability & Proof of Execution**: Ensure execution produces tangible verification signals (terminal output, tests, metrics).
+
+## Activation Boundaries
+
+- **Activate when:** Use when implementing WCAG 2.2 AA compliant UI motion, tiered reduced-motion patterns, vestibular trigger suppression, and accessible transitions across CSS, React, Framer Motion, and GSAP.
+- **DO NOT activate when:** The task falls outside the `accessible-animation` domain or is managed by a different dedicated specialist.
+
+## 🔁 Multi-Pass Execution Protocol
+
+Execute all non-trivial tasks through this 7-pass cognitive loop:
+
+| Pass | Phase | Core Action |
+|:---|:---|:---|
+| **Pass 1** | **Understand** | Deconstruct the user's explicit objective, implicit requirements, and platform constraints. |
+| **Pass 2** | **Plan** | Decompose the task into smallest logical steps; map dependencies and required tool calls. |
+| **Pass 3** | **Execute** | Implement the solution with production-grade craft, zero placeholders, and strict typing. |
+| **Pass 4** | **Verify** | Run linters, unit tests, or compiler checks to validate structural correctness. |
+| **Pass 5** | **Attack** | Perform an adversarial review searching for edge-case failures, race conditions, and traps. |
+| **Pass 6** | **Improve** | Eliminate discovered friction, optimize performance, and harden error boundaries. |
+| **Pass 7** | **Quality Gate** | Enforce Verification-Before-Completion (VBC) with concrete terminal proof before finalizing. |
 
 ---
 
-## Mandatory Pre-Flight Context Inspection
+## 🛠️ Technical Architecture & Reference Recipes
 
-Before implementing UI animation code, you MUST inspect:
+Motion in digital interfaces elevates feedback and delight, but for users with vestibular disorders, inner-ear sensitivities, or cognitive processing differences, unrestrained motion causes severe nausea, vertigo, migraines, and physical disorientation. A master-grade accessible motion architecture does not bluntly disable all UI feedback; it systematically applies tiered reduction—preserving vital state changes and subtle opacity fades while eliminating spatial translations, zoom scales, parallax shifts, and rapid flashes.
 
-1. CSS stylesheets → Verify presence of `@media (prefers-reduced-motion: reduce)` rules
-2. Framer Motion / GSAP components → Query `useReducedMotion()` or `matchMedia('(prefers-reduced-motion: reduce)')`
-3. Vestibular Trigger Replacement (Section 57) → Replace parallax scroll and 3D rotations with instant opacity cross-fades (`0 -> 1`)
+---
 
-Ensure UI motion respects user accessibility preferences (`prefers-reduced-motion: reduce`) without stripping functional state updates.
 
 ---
 
-## 3 Tiered Reduced-Motion Rules
 
-### Tier 1: CSS Reduced Motion Media Query
+---
+
+
+---
+
+## 🛠️ Technical Architecture & Reference Recipes
+
+### 1. The Vestibular Trigger Substitution Matrix
+
+| Disorienting Trigger | Vestibular Hazard | Accessible Compliant Substitution |
+|:---|:---|:---|
+| **Full-Page Parallax** | Severe nausea, spatial imbalance | Static layout positioning with subtle content opacity reveal |
+| **Multi-Axis 3D Rotation** | Visual vertigo, disorientation | 2D elevation shadow transition or subtle border glow |
+| **Rapid Zoom / Scale (>1.2x)** | Depth perception distortion | Gentle scale (1.0 to 1.02x) or immediate 150ms opacity cross-fade |
+| **Infinite Spinning Spinners** | Cognitive fatigue, dizziness | Pulsing opacity indicator or static step progress bar |
+| **Off-Canvas Fly-In Drawer** | Visual lurching across viewport | Immediate opacity dissolve (`0` to `1` over 120ms) |
+
+---
+
+### 2. Four Production-Grade Implementation Recipes
+
+#### Recipe 1: CSS Non-Destructive Global Safety Harness
 
 ```css
+/* Universal Non-Destructive Reset: Preserves JS event hooks while eliminating movement */
 @media (prefers-reduced-motion: reduce) {
   *,
-  ::before,
-  ::after {
-    animation-duration: 0.01ms !important;
+  *::before,
+  *::after {
+    /* 0.001ms duration ensures transitionend & animationend events still fire reliably */
+    animation-duration: 0.001ms !important;
     animation-iteration-count: 1 !important;
-    transition-duration: 0.01ms !important;
+    transition-duration: 0.001ms !important;
     scroll-behavior: auto !important;
   }
+
+  /* Exceptions: Allow intentional, gentle micro-fades for essential state comprehension */
+  .accessible-fade {
+    transition: opacity 150ms ease-out !important;
+  }
 }
 ```
 
-### Tier 2: React & Framer Motion Hook (`useReducedMotion`)
+---
+
+#### Recipe 2: SSR-Safe React Hook (`useAccessibleMotion`)
+
+Prevents hydration mismatch warnings by defaulting to safe state until client mounts:
 
 ```tsx
-import { useReducedMotion, motion } from 'framer-motion';
+import { useState, useEffect } from 'react';
+
+export function useAccessibleMotion(): boolean {
+  // Default to false on server to match initial HTML render
+  const [shouldReduceMotion, setShouldReduceMotion] = useState(false);
+
+  useEffect(() => {
+    if (typeof window === 'undefined' || !window.matchMedia) return;
 
-export function AccessibleCard({ children }) {
-  const shouldReduceMotion = useReducedMotion();
+    const mediaQuery = window.matchMedia('(prefers-reduced-motion: reduce)');
+    setShouldReduceMotion(mediaQuery.matches);
+
+    const handleChange = (e: MediaQueryListEvent) => {
+      setShouldReduceMotion(e.matches);
+    };
+
+    mediaQuery.addEventListener('change', handleChange);
+    return () => mediaQuery.removeEventListener('change', handleChange);
+  }, []);
+
+  return shouldReduceMotion;
+}
+```
+
+---
+
+#### Recipe 3: Framer Motion Tiered Component Pattern
+
+```tsx
+import React from 'react';
+import { motion, useReducedMotion } from 'framer-motion';
+
+interface AccessibleDialogProps {
+  isOpen: boolean;
+  onClose: () => void;
+  children: React.ReactNode;
+}
+
+export function AccessibleDialog({ isOpen, onClose, children }: AccessibleDialogProps) {
+  const shouldReduce = useReducedMotion();
+
+  if (!isOpen) return null;
 
   return (
-    <motion.div
-      initial={{ opacity: 0, y: shouldReduceMotion ? 0 : 20 }}
-      animate={{ opacity: 1, y: 0 }}
-      transition={{ duration: shouldReduceMotion ? 0.05 : 0.3 }}
-    >
-      {children}
-    </motion.div>
+    <div className="fixed inset-0 z-50 flex items-center justify-center bg-black/60">
+      <motion.div
+        role="dialog"
+        aria-modal="true"
+        initial={shouldReduce ? { opacity: 0 } : { opacity: 0, scale: 0.92, y: 16 }}
+        animate={{ opacity: 1, scale: 1, y: 0 }}
+        exit={shouldReduce ? { opacity: 0 } : { opacity: 0, scale: 0.94, y: 12 }}
+        transition={{
+          duration: shouldReduce ? 0.12 : 0.22,
+          ease: [0.16, 1, 0.3, 1],
+        }}
+        className="w-full max-w-lg p-6 bg-white dark:bg-neutral-900 rounded-2xl shadow-2xl"
+      >
+        {children}
+      </motion.div>
+    </div>
   );
 }
 ```
 
-### Tier 3: Replacing Parallax & Vestibular Triggers with Instant Fades
+---
+
+#### Recipe 4: GSAP `matchMedia` Integration & Smooth-Scroll Cancellation
+
+```typescript
+import gsap from 'gsap';
+import { ScrollTrigger } from 'gsap/ScrollTrigger';
+
+gsap.registerPlugin(ScrollTrigger);
+
+export function setupAccessibleScrollMotion(targetElement: HTMLElement) {
+  const mm = gsap.matchMedia();
+
+  // 1. Standard Motion (User has not requested reduction)
+  mm.add('(prefers-reduced-motion: no-preference)', () => {
+    gsap.to(targetElement, {
+      y: -80,
+      scale: 1.05,
+      scrollTrigger: {
+        trigger: targetElement,
+        start: 'top bottom',
+        end: 'bottom top',
+        scrub: 1,
+      },
+    });
+  });
+
+  // 2. Reduced Motion (User prefers reduced motion)
+  mm.add('(prefers-reduced-motion: reduce)', () => {
+    // Zero parallax scrub, zero scale shift. Pure gentle opacity cross-fade.
+    gsap.fromTo(
+      targetElement,
+      { opacity: 0.4 },
+      {
+        opacity: 1,
+        duration: 0.2,
+        scrollTrigger: {
+          trigger: targetElement,
+          start: 'top 85%',
+          toggleActions: 'play none none reverse',
+        },
+      }
+    );
+  });
 
-- Disorienting motions (parallax scrolling, 3D rotations, zoom scaling) MUST be converted into instant cross-fades (`opacity: 0 -> 1`) when `prefers-reduced-motion: reduce` is active.
+  return () => mm.revert();
+}
+```
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🚨 Edge-Case & Failure Mode Matrix
 
-1. **Stripping All State Updates**: Completely disabling CSS transitions so elements abruptly disappear or jump without opacity fading.
-2. **Ignoring JS Smooth Scroll**: Leaving smooth scroll libraries (Lenis, GSAP ScrollSmoother) active when reduced motion is preferred.
+| Scenario | Risk | Mitigation Strategy |
+|:---|:---|:---|
+| **Empty or Null Inputs** | Unhandled exception or unexpected rendering collapse | Enforce fallback guards, optional chaining, and explicit empty state handlers |
+| **Network Timeout / Latency** | Hanging operations or duplicate side-effects | Implement bounded abort controllers, exponential backoff, and idempotency keys |
+| **Concurrency / Race Conditions** | Stale state overwrite or inconsistent data mutations | Use atomic transactions, mutex locking, or cancel-on-resubmit controls |
+| **Invalid Schema / Malformed Payload** | Downstream runtime errors or security injection | Validate boundary payloads with Zod/Pydantic schemas prior to execution |
+| **Resource / Memory Saturation** | OOM errors, frame drops, or memory leaks | Clean up listeners, cancel active timers, and enforce pagination/virtualization |
 
 ---
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
+## 🤖 LLM-Specific Traps Table
 
-**Active reviewers: `accessibility-reviewer` · `motion-reviewer`**
+| Anti-Pattern | What AI Commonly Does Wrong | What Is Actually Correct |
+|:---|:---|:---|
+| **The Instant Pop Trap** | Conditionally unmounting elements without animated interpolation | Use AnimatePresence or coordinate morphs with continuous geometry |
+| **Layout Thrashing** | Animating width, height, top, or left inside animation loops | Animate composite-only transform (translate3d, scale) and opacity |
+| **Sluggish Duration** | Setting micro-interaction transitions to 600ms+ causing interface lag | Cap interactive feedback at 160ms–240ms with snappy ease-out curves |
+
+---
 
-### ✅ Pre-Flight Self-Audit
+## 🏛️ Tribunal Verification & Guardrails
+
+**Active Reviewers:** `frontend-reviewer` · `motion-reviewer` · `ui-ux-auditor`
+**Slash Command:** `/review` or `/tribunal-full`
+
+### ✅ Pre-Flight Self-Audit Checklist
 
 ```
-✅ Is `@media (prefers-reduced-motion: reduce)` declared in CSS stylesheets?
-✅ Do JS animation libraries (Framer Motion / GSAP) query `useReducedMotion()` or `window.matchMedia`?
-✅ Are disorienting parallax/zoom effects converted into gentle opacity fades?
+✅ Does animation maintain 60fps/120fps using transform and opacity?
+✅ Is optical mass conserved across state interpolations without volume collapse?
+✅ Is duration capped within micro-interaction budgets (160ms–280ms)?
+✅ Is prefers-reduced-motion respected with graceful instant fallbacks?
+✅ Did I prevent layout thrashing and continuous geometry mutations?
 ```
 
 ### 🛑 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.
+- ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing test suites, compiler success, or equivalent operational proof) that your output works as intended.

package/.agent/skills/adapt/SKILL.md

@@ -1,8 +1,8 @@
 ---
 name: adapt
-description: Adapt designs across breakpoints, devices, platform constraints, touch vs mouse input, and container queries. Use when making a UI responsive or optimizing for mobile/tablet.
-version: 3.0.0
-last-updated: 2026-07-30
+description: Use when Adapt designs across breakpoints, devices, platform constraints, touch vs mouse input, and container queries. Use when making a UI responsive or optimizing for mobile/tablet.
+version: 5.0.0
+last-updated: 2026-09-13
 skills:
   - mobile-design
   - building-native-ui
@@ -15,17 +15,41 @@ scripts-binding:
 
 # Adapt — Responsive Adaptation & Container Queries
 
----
-
 ## Mandatory Pre-Flight Context Inspection
 
-Before engineering responsive layouts or cross-device UIs, you MUST inspect:
+Before generating, refactoring, or reviewing code in the `adapt` domain, inspect these 5 critical parameters:
+
+1. **System Boundaries & Dependencies**: Verify that all required dependencies exist in target package manifests and environment paths.
+2. **Runtime Context & Platform Invariants**: Confirm target platform constraints (Node.js, Browser, Mobile OS, Edge runtime) before applying APIs.
+3. **Execution Guardrails**: Identify potential side-effects, state mutations, and unhandled asynchronous exceptions.
+4. **Validation & Type Contracts**: Validate input data schemas and strict type constraints across all module interfaces.
+5. **Observability & Proof of Execution**: Ensure execution produces tangible verification signals (terminal output, tests, metrics).
+
+## Activation Boundaries
+
+- **Activate when:** Use when Adapt designs across breakpoints, devices, platform constraints, touch vs mouse input, and container queries. Use when making a UI responsive or optimizing for mobile/tablet.
+- **DO NOT activate when:** The task falls outside the `adapt` domain or is managed by a different dedicated specialist.
 
-1. Container Queries (`@container`) over Viewport Media Queries (Section 24) → Adapt component layouts based on parent container width (`container-type: inline-size`)
-2. Touch Target Sizing (Section 39) → Ensure interactive controls meet min `44x44px` target size on touch devices
-3. Fluid Clamp Scaling (Section 45) → Use `clamp()` for smooth fluid typography and spacing without abrupt media query jumps
+## 🔁 Multi-Pass Execution Protocol
+
+Execute all non-trivial tasks through this 7-pass cognitive loop:
+
+| Pass | Phase | Core Action |
+|:---|:---|:---|
+| **Pass 1** | **Understand** | Deconstruct the user's explicit objective, implicit requirements, and platform constraints. |
+| **Pass 2** | **Plan** | Decompose the task into smallest logical steps; map dependencies and required tool calls. |
+| **Pass 3** | **Execute** | Implement the solution with production-grade craft, zero placeholders, and strict typing. |
+| **Pass 4** | **Verify** | Run linters, unit tests, or compiler checks to validate structural correctness. |
+| **Pass 5** | **Attack** | Perform an adversarial review searching for edge-case failures, race conditions, and traps. |
+| **Pass 6** | **Improve** | Eliminate discovered friction, optimize performance, and harden error boundaries. |
+| **Pass 7** | **Quality Gate** | Enforce Verification-Before-Completion (VBC) with concrete terminal proof before finalizing. |
+
+---
+
+## 🛠️ Technical Architecture & Reference Recipes
+
+---
 
-Architect UIs that adapt fluidly to screen dimensions, container boundaries, input devices (touch vs pointer), and orientation.
 
 ---
 
@@ -63,28 +87,45 @@ Architect UIs that adapt fluidly to screen dimensions, container boundaries, inp
 
 ---
 
-## 🤖 LLM-Specific Traps
+## 🚨 Edge-Case & Failure Mode Matrix
+
+| Scenario | Risk | Mitigation Strategy |
+|:---|:---|:---|
+| **Empty or Null Inputs** | Unhandled exception or unexpected rendering collapse | Enforce fallback guards, optional chaining, and explicit empty state handlers |
+| **Network Timeout / Latency** | Hanging operations or duplicate side-effects | Implement bounded abort controllers, exponential backoff, and idempotency keys |
+| **Concurrency / Race Conditions** | Stale state overwrite or inconsistent data mutations | Use atomic transactions, mutex locking, or cancel-on-resubmit controls |
+| **Invalid Schema / Malformed Payload** | Downstream runtime errors or security injection | Validate boundary payloads with Zod/Pydantic schemas prior to execution |
+| **Resource / Memory Saturation** | OOM errors, frame drops, or memory leaks | Clean up listeners, cancel active timers, and enforce pagination/virtualization |
+
+---
+
+## 🤖 LLM-Specific Traps Table
 
-1. **Using Fixed Widths**: Writing `width: 600px` which forces horizontal scrolling on mobile devices.
-2. **Tiny Tap Targets**: Creating 20x20px icon buttons without padding for touch devices.
+| Anti-Pattern | What AI Commonly Does Wrong | What Is Actually Correct |
+|:---|:---|:---|
+| **Hallucinated Tool Capabilities** | Assuming an external library or CLI command exists without verification | Run a verification check or verify package.json before referencing tools |
+| **Premature Completion Claim** | Declaring a task finished because code was generated without verification | Execute tests, linters, or terminal commands to provide concrete proof |
+| **Context Bloat Dumping** | Pasting entire multi-thousand-line files into prompt context | Extract targeted excerpts, symbols, and signatures to preserve tokens |
 
 ---
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
+## 🏛️ Tribunal Verification & Guardrails
 
-**Active reviewers: `frontend-reviewer` · `mobile-reviewer`**
+**Active Reviewers:** `orchestrator` · `agent-organizer` · `logic-reviewer`
+**Slash Command:** `/review` or `/tribunal-full`
 
-### ✅ Pre-Flight Self-Audit
+### ✅ Pre-Flight Self-Audit Checklist
 
 ```
-✅ Are re-usable components using `@container` queries for layout adaptation?
-✅ Do all touch interactive targets meet the 44x44px minimum sizing rule?
-✅ Does the layout prevent horizontal scrollbars on 320px screen widths?
+✅ Did I deconstruct the root objective before proposing architecture?
+✅ Did I identify dependencies, bottlenecks, and parallelizable sub-tasks?
+✅ Did I avoid over-engineering and select the simplest effective pattern?
+✅ Did I verify assumptions with concrete file reads instead of speculation?
+✅ Did I establish measurable verification criteria before completion?
 ```
 
 ### 🛑 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.
+- ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing test suites, compiler success, or equivalent operational proof) that your output works as intended.

package/.agent/skills/advanced-rag-pipelines/SKILL.md

@@ -1,13 +1,13 @@
 ---
 name: advanced-rag-pipelines
-description: Production-grade Retrieval-Augmented Generation (RAG) mastery. Semantic chunking, Hybrid Search (Dense + Sparse/BM25), Cross-Encoder Reranking, and architecture-agnostic vector database management.
-tools: Read, Grep, Glob, Bash, Edit, Write
-version: 3.0.0
-last-updated: 2026-07-30
+description: Use when Production-grade Retrieval-Augmented Generation (RAG) mastery. Semantic chunking, Hybrid Search (Dense + Sparse/BM25), Cross-Encoder Reranking, and architecture-agnostic vector database management.
+version: 5.0.0
+last-updated: 2026-09-13
 skills:
   - llm-engineering
   - ai-prompt-injection-defense
   - database-design
+tools: Read, Grep, Glob, Bash, Edit, Write
 scripts-binding:
   - .agent/scripts/lint_runner.js
   - .agent/scripts/verify_all.js
@@ -15,25 +15,71 @@ scripts-binding:
 
 # Advanced RAG Pipelines (Production AI Data)
 
----
-
 ## Mandatory Pre-Flight Context Inspection
 
-Before building RAG pipelines or vector search components, you MUST inspect:
+Before generating, refactoring, or reviewing code in the `advanced-rag-pipelines` domain, inspect these 5 critical parameters:
 
-1. Hybrid Search Requirement (Section 16) → Always combine Dense Vector Search with Sparse BM25 Search to catch exact keyword matches (IDs, versions)
-2. Two-Stage Reranking Pipeline (Section 29) → Retrieve top ~50 candidate chunks, then rerank with a Cross-Encoder down to top 3-5 before feeding the LLM
-3. Context XML Framing (Section 59) → Wrap retrieved chunks inside explicit `<context>` XML tags in the prompt to prevent indirect prompt injection
+1. **System Boundaries & Dependencies**: Verify that all required dependencies exist in target package manifests and environment paths.
+2. **Runtime Context & Platform Invariants**: Confirm target platform constraints (Node.js, Browser, Mobile OS, Edge runtime) before applying APIs.
+3. **Execution Guardrails**: Identify potential side-effects, state mutations, and unhandled asynchronous exceptions.
+4. **Validation & Type Contracts**: Validate input data schemas and strict type constraints across all module interfaces.
+5. **Observability & Proof of Execution**: Ensure execution produces tangible verification signals (terminal output, tests, metrics).
 
-# Advanced RAG Pipelines (Production AI Data)
+## Activation Boundaries
+
+- **Activate when:** Use when Production-grade Retrieval-Augmented Generation (RAG) mastery. Semantic chunking, Hybrid Search (Dense + Sparse/BM25), Cross-Encoder Reranking, and architecture-agnostic vector database management.
+- **DO NOT activate when:** The task falls outside the `advanced-rag-pipelines` domain or is managed by a different dedicated specialist.
+
+## 🔁 Multi-Pass Execution Protocol
+
+Execute all non-trivial tasks through this 7-pass cognitive loop:
+
+| Pass | Phase | Core Action |
+|:---|:---|:---|
+| **Pass 1** | **Understand** | Deconstruct the user's explicit objective, implicit requirements, and platform constraints. |
+| **Pass 2** | **Plan** | Decompose the task into smallest logical steps; map dependencies and required tool calls. |
+| **Pass 3** | **Execute** | Implement the solution with production-grade craft, zero placeholders, and strict typing. |
+| **Pass 4** | **Verify** | Run linters, unit tests, or compiler checks to validate structural correctness. |
+| **Pass 5** | **Attack** | Perform an adversarial review searching for edge-case failures, race conditions, and traps. |
+| **Pass 6** | **Improve** | Eliminate discovered friction, optimize performance, and harden error boundaries. |
+| **Pass 7** | **Quality Gate** | Enforce Verification-Before-Completion (VBC) with concrete terminal proof before finalizing. |
+
+---
+
+## 🛠️ Technical Architecture & Reference Recipes
+
+---
 
-You are an expert in building production-grade Retrieval-Augmented Generation (RAG) data pipelines. You understand that naive RAG (fixed chunking + Cosine similarity) fails in production. You architect systems that retrieve context with high precision using hybrid search, reranking, and semantic strategies.
 
-## 1. Core Principles
+## 2026 RAG Performance & Vector Invariants
+
+1. **Reciprocal Rank Fusion (RRF)**:
+   ```python
+   # Combine dense + sparse rankings without normalizing disparate score distributions
+   def rrf(dense_ranks: dict[str, int], sparse_ranks: dict[str, int], k: int = 60) -> dict[str, float]:
+       scores = {}
+       for doc_id, rank in dense_ranks.items():
+           scores[doc_id] = scores.get(doc_id, 0) + 1 / (k + rank)
+       for doc_id, rank in sparse_ranks.items():
+           scores[doc_id] = scores.get(doc_id, 0) + 1 / (k + rank)
+       return dict(sorted(scores.items(), key=lambda x: x[1], reverse=True))
+   ```
+2. **HNSW Indexing with Halfvec (pgvector 0.8+)**:
+   ```sql
+   -- Halves memory usage with < 1% recall loss
+   CREATE INDEX idx_docs_embedding ON documents
+   USING hnsw ((embedding::halfvec(1536)) halfvec_cosine_ops);
+   ```
+3. **Aggressive Context Pruning**: Never dump > 5 chunks into the final LLM prompt. Context dilution ("Lost in the Middle") degrades factual recall and spikes latency.
+
+## Hallucination Traps (Read First)
+
+- ❌ Fixed-character chunking (e.g. split every 500 chars) → ✅ AST/Markdown-aware structural chunking
+- ❌ Relying only on cosine similarity on raw queries → ✅ Hybrid search (Dense + BM25) with cross-encoder rerank
+- ❌ Injecting raw text into system prompt → ✅ Enclose in `<retrieved_context>` tags to prevent indirect prompt injection
+- ❌ Full precision FP32 vectors on massive datasets → ✅ Use FP16 (`halfvec`) or scalar quantization
 
-- **Garbage In, Garbage Out:** Vector embeddings are only as good as the chunking strategy. Never use arbitrary character counts for chunking code or complex documents.
-- **Hybrid Search is Mandatory:** Dense vectors (embeddings) are terrible at exact keyword matches (e.g., finding "ID-4912" or "v4.4.4"). Always combine Dense Search with Sparse Search (BM25) to catch both semantic intent and exact matches.
-- **Retrieve Many, Rerank to Few:** It is cheaper and more accurate to retrieve 50 candidate chunks from a Vector DB and use a Cross-Encoder to rerank them down to the top 5 for the LLM.
+---
 
 ## 2. Advanced Architectural Patterns
 
@@ -77,52 +123,47 @@ Before submitting code, ensure:
 2. BM25 / Sparse search is considered alongside standard dense embeddings.
 3. Chunks are injected into the final LLM prompt with explicit `<context>` XML boundaries to prevent prompt injection.
 
-### 🛑 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.
-
 ---
 
-## 🤖 LLM-Specific Traps
+## 🚨 Edge-Case & Failure Mode Matrix
 
-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.
+| Scenario | Risk | Mitigation Strategy |
+|:---|:---|:---|
+| **Empty or Null Inputs** | Unhandled exception or unexpected rendering collapse | Enforce fallback guards, optional chaining, and explicit empty state handlers |
+| **Network Timeout / Latency** | Hanging operations or duplicate side-effects | Implement bounded abort controllers, exponential backoff, and idempotency keys |
+| **Concurrency / Race Conditions** | Stale state overwrite or inconsistent data mutations | Use atomic transactions, mutex locking, or cancel-on-resubmit controls |
+| **Invalid Schema / Malformed Payload** | Downstream runtime errors or security injection | Validate boundary payloads with Zod/Pydantic schemas prior to execution |
+| **Resource / Memory Saturation** | OOM errors, frame drops, or memory leaks | Clean up listeners, cancel active timers, and enforce pagination/virtualization |
 
 ---
 
-## 🏛️ Tribunal Integration (Anti-Hallucination)
+## 🤖 LLM-Specific Traps Table
 
-**Slash command: `/review` or `/tribunal-full`**
-**Active reviewers: `logic-reviewer` · `security-auditor`**
+| Anti-Pattern | What AI Commonly Does Wrong | What Is Actually Correct |
+|:---|:---|:---|
+| **Unchecked Payload Cast** | Casting request bodies to TypeScript types without runtime schema validation | Parse request payloads through Zod/Pydantic schemas before business logic |
+| **Silent Error Swallowing** | Catching errors with empty catch blocks or logging without rethrowing | Propagate structured errors with status codes and contextual stack traces |
+| **Unparameterized Query** | Concatenating user inputs into SQL/Prisma query strings | Always use parameterized bindings or type-safe ORM query builders |
 
-### ❌ 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.
+## 🏛️ Tribunal Verification & Guardrails
 
-### ✅ Pre-Flight Self-Audit
+**Active Reviewers:** `logic-reviewer` · `security-auditor` · `api-architect` · `resilience-reviewer`
+**Slash Command:** `/review` or `/tribunal-full`
 
-Review these questions before confirming output:
+### ✅ Pre-Flight Self-Audit Checklist
 
 ```
-✅ 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?
+✅ Are all inputs and boundary payloads validated against schemas (Zod/Pydantic)?
+✅ Are SQL and database queries parameterized with zero string concatenation?
+✅ Are error boundaries and timeout/retry policies explicitly declared?
+✅ Are authentication checks performed before business logic execution?
+✅ Did I verify that imported dependencies exist in package.json/requirements.txt?
 ```
 
 ### 🛑 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.
+- ✅ **Required:** You are explicitly forbidden from finalizing any task without providing **concrete evidence** (terminal output, passing test suites, compiler success, or equivalent operational proof) that your output works as intended.