tribunal-kit 4.5.1 → 4.6.1

package/.agent/skills/generative-ui-expert/SKILL.md

@@ -1,6 +1,9 @@
 ---
 name: generative-ui-expert
 description: Generative UI mastery. Vercel AI SDK 3.0+, React Server Components (RSC) + LLMs, streaming UI elements, structured tool calling (Zod schemas), and managing client-side AI state via useChat/useObject.
+routing:
+  domain: general
+  tier: basic
 ---
 
 # Generative UI Expert (Vercel AI SDK)
@@ -8,6 +11,7 @@ description: Generative UI mastery. Vercel AI SDK 3.0+, React Server Components
 You are the definitive expert in Generative UI using the Vercel AI SDK and React Server Components (RSC). Your goal is to move AI from spitting out "markdown walls of text" into rendering interactive, stateful, and dynamic UI components natively inside the chat or application stream.
 
 ## 1. Core Principles
+
 - **No Markdown Slop:** Avoid dumping raw markdown when structured UI can be used. If the user asks for a weather report, stream a `<WeatherCard />`, not text.
 - **Server-Driven UI:** Leverage React Server Components (`ai/rsc`) to stream actual React components over the wire as the LLM yields function calls.
 - **Structured Data First:** Use strict Zod schemas (`useObject`, `streamObject`) whenever you need the LLM to output parsable data.
@@ -16,7 +20,9 @@ You are the definitive expert in Generative UI using the Vercel AI SDK and React
 ## 2. Vercel AI SDK Patterns
 
 ### A. Streaming React Components (`ai/rsc`)
+
 When setting up `ai/rsc`, define explicit tool boundaries:
+
 ```typescript
 import { createAI, getMutableAIState, streamUI } from "ai/rsc";
 import { z } from "zod";
@@ -47,12 +53,14 @@ export const AI = createAI({
 ```
 
 ### B. Structured Output (`streamObject`)
+
 Use this when you need strict JSON streams for charts, tables, or complex states.
+
 ```typescript
 const result = await streamObject({
   model: openai("gpt-4-turbo"),
   schema: z.object({
-    points: z.array(z.object({ x: z.number(), y: z.number() }))
+    points: z.array(z.object({ x: z.number(), y: z.number() })),
   }),
   prompt: "Generate a sales forecast chart data",
 });
@@ -60,18 +68,22 @@ const result = await streamObject({
 ```
 
 ## 3. Client-Side State Management
+
 - Use `useChat` for standard text+tool workflows.
 - Use `useUIState` and `useAIState` to manage the UI payload array and the underlying LLM message history separately.
 - Always include `id` and `role` in message schemas to prevent key-rendering bugs in React.
 
 ## 4. LLM Traps & Pre-Flight Checks
+
 - **TRAP:** Sending client components directly over the wire from `generate:`.
 - **FIX:** Server actions can only return Server Components. If returning an interactive widget, wrap it in a client component but yield it from the server.
 - **TRAP:** Forgetting to yield intermediate states in slow tools.
 - **FIX:** Always `yield <Loading />` before awaiting slow API calls inside a tool's `generate` function.
 
 ## Verification Protocol
+
 Before submitting code, ensure:
+
 1. `zod` is used for all tool parameters.
 2. Server Actions are properly annotated with `"use server"`.
 3. The model supports tool calling (e.g., `gpt-4o`, `claude-3-5-sonnet`).
@@ -79,5 +91,49 @@ Before submitting code, ensure:
 ### 🛑 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
+
+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)
+
+**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?
+✅ 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/geo-fundamentals/SKILL.md

@@ -1,151 +1,150 @@
----
-name: geo-fundamentals
-description: Generative Engine Optimization (GEO) mastery. Structuring content for LLM ingestion. Adapting SEO for AI interfaces (ChatGPT, Claude, Perplexity), optimizing markdown semantic hierarchies, citation structuring, minimizing boilerplate, reducing HTML DOM depth, and API-first content delivery. Use when making information discoverable not just to Google, but directly to AI indexing agents.
-allowed-tools: Read, Write, Edit, Glob, Grep
-version: 2.0.0
-last-updated: 2026-04-02
-applies-to-model: gemini-2.5-pro, claude-3-7-sonnet
----
-
-## Hallucination Traps (Read First)
-- ❌ Assuming SEO and GEO are the same -> ✅ GEO optimizes for AI crawlers and LLM ingestion, not just search engine ranking
-- ❌ Using complex nested HTML for content -> ✅ AI parsers prefer flat, semantic HTML with clear heading hierarchies
-- ❌ Ignoring structured data (JSON-LD) -> ✅ LLMs heavily weight structured data for citation and fact extraction
-
----
-
-
-# Generative Engine Optimization (GEO) 
-
----
-
-## 1. The Death of Boilerplate (Information Density)
-
-When ChatGPT or Perplexity queries a website, it has a finite context window. 
-
-If your website contains 8,000 words of "fluff" marketing copy and only 200 words of actionable data (pricing, API limits, support contact), the LLM will truncate the page and hallucinates the rest.
-
-### The GEO Markdown Fallback
-Modern sites should natively serve structured markdown if they detect an AI User-Agent (like `ChatGPT-User` or `PerplexityBot`).
-
-```typescript
-// Next.js Edge Middleware for GEO
-export function middleware(req: NextRequest) {
-  const ua = req.headers.get('user-agent') || '';
-  const isBot = /ChatGPT|Perplexity|ClaudeBot/i.test(ua);
-
-  if (isBot) {
-    // Reroute the AI bot to a hyper-dense, unstyled Markdown data dump
-    // This removes 3MB of React DOM hierarchy, getting straight to the facts.
-    return NextResponse.rewrite(new URL(`/api/geo-export${req.nextUrl.pathname}`, req.url));
-  }
-}
-```
-
----
-
-## 2. Citation Optimization
-
-LLMs (like Perplexity and Gemini Search) require explicit sources to reference your website in their output.
-If a statistic on your site is hard to source, the AI will ignore it in favor of a competitor.
-
-**Rule: Explicit Claim-to-Source Mapping**
-Do not use vague external links at the bottom of the page. Anchor exact claims to explicit references directly within the text block. Use standardized `<cite>` tags or JSON-LD.
-
-```html
-<!-- ❌ BAD: Vague SEO -->
-<p>We are the fastest vector database on the market. Read our docs.</p>
-
-<!-- ✅ GOOD: GEO Citation Architecture -->
-<p>We process 10M vectors at 10ms latency (P99), making us 3x faster than Competitor A.</p>
-<cite xmlns="http://schema.org" typeof="WebPage">
-  <span property="name">Benchmark Methodology 2026</span> - 
-  <a property="url" href="/benchmarks-2026.pdf">[Source PDF]</a>
-</cite>
-```
-
----
-
-## 3. High-Clarity Semantic Taxonomy
-
-Traditional SEO relies heavily on long-tail keyword placement.
-GEO relies heavily on **Taxonomy and Relational Mapping**. The LLM wants to know exactly *what* entity this page represents.
-
-**Implement Explicit FAQs:**
-LLMs love QA formats because user prompts are usually questions.
-Transform prose into rigid QA objects using standard semantic data blocks.
-
-```html
-<dl>
-  <dt><h3>What is the data retention limit for the Free Tier?</h3></dt>
-  <dd><p>The Free Tier limits data retention to precisely 14 rolling days.</p></dd>
-  
-  <dt><h3>Does the platform support HIPAA compliance?</h3></dt>
-  <dd><p>Yes, Enterprise Tiers support full BAA HIPAA compliance parameters.</p></dd>
-</dl>
-```
-
----
-
-## 4. API Docs Readability (The Primary AI Target)
-
-When an AI tries to write code using your product, it scrapes your API documentation.
-
-1. **Eliminate Image-Based Architecture:** Highlighting your architecture solely in a PNG graphic is invisible. Describe the system architecture using text or standard Mermaid.js code blocks.
-2. **Provide Concrete Copy-Paste Examples:** The easiest way to get an LLM to use your platform correctly is to ensure your docs contain perfectly working `curl`, `TypeScript`, and `Python` code snippets with zero external dependencies.
-
----
-
-
----
-
-
-
-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.
-
----
-
-
-
-**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.
+---
+name: geo-fundamentals
+description: Generative Engine Optimization (GEO) mastery. Structuring content for LLM ingestion. Adapting SEO for AI interfaces (ChatGPT, Claude, Perplexity), optimizing markdown semantic hierarchies, citation structuring, minimizing boilerplate, reducing HTML DOM depth, and API-first content delivery. Use when making information discoverable not just to Google, but directly to AI indexing agents.
+allowed-tools: Read, Write, Edit, Glob, Grep
+version: 2.0.0
+last-updated: 2026-04-02
+applies-to-model: gemini-2.5-pro, claude-3-7-sonnet
+routing:
+  domain: general
+  tier: basic
+---
+
+## Hallucination Traps (Read First)
+
+- ❌ Assuming SEO and GEO are the same -> ✅ GEO optimizes for AI crawlers and LLM ingestion, not just search engine ranking
+- ❌ Using complex nested HTML for content -> ✅ AI parsers prefer flat, semantic HTML with clear heading hierarchies
+- ❌ Ignoring structured data (JSON-LD) -> ✅ LLMs heavily weight structured data for citation and fact extraction
+
+---
+
+# Generative Engine Optimization (GEO)
+
+---
+
+## 1. The Death of Boilerplate (Information Density)
+
+When ChatGPT or Perplexity queries a website, it has a finite context window.
+
+If your website contains 8,000 words of "fluff" marketing copy and only 200 words of actionable data (pricing, API limits, support contact), the LLM will truncate the page and hallucinates the rest.
+
+### The GEO Markdown Fallback
+
+Modern sites should natively serve structured markdown if they detect an AI User-Agent (like `ChatGPT-User` or `PerplexityBot`).
+
+```typescript
+// Next.js Edge Middleware for GEO
+export function middleware(req: NextRequest) {
+  const ua = req.headers.get("user-agent") || "";
+  const isBot = /ChatGPT|Perplexity|ClaudeBot/i.test(ua);
+
+  if (isBot) {
+    // Reroute the AI bot to a hyper-dense, unstyled Markdown data dump
+    // This removes 3MB of React DOM hierarchy, getting straight to the facts.
+    return NextResponse.rewrite(new URL(`/api/geo-export${req.nextUrl.pathname}`, req.url));
+  }
+}
+```
+
+---
+
+## 2. Citation Optimization
+
+LLMs (like Perplexity and Gemini Search) require explicit sources to reference your website in their output.
+If a statistic on your site is hard to source, the AI will ignore it in favor of a competitor.
+
+**Rule: Explicit Claim-to-Source Mapping**
+Do not use vague external links at the bottom of the page. Anchor exact claims to explicit references directly within the text block. Use standardized `<cite>` tags or JSON-LD.
+
+```html
+<!-- ❌ BAD: Vague SEO -->
+<p>We are the fastest vector database on the market. Read our docs.</p>
+
+<!-- ✅ GOOD: GEO Citation Architecture -->
+<p>We process 10M vectors at 10ms latency (P99), making us 3x faster than Competitor A.</p>
+<cite xmlns="http://schema.org" typeof="WebPage">
+  <span property="name">Benchmark Methodology 2026</span> -
+  <a property="url" href="/benchmarks-2026.pdf">[Source PDF]</a>
+</cite>
+```
+
+---
+
+## 3. High-Clarity Semantic Taxonomy
+
+Traditional SEO relies heavily on long-tail keyword placement.
+GEO relies heavily on **Taxonomy and Relational Mapping**. The LLM wants to know exactly _what_ entity this page represents.
+
+**Implement Explicit FAQs:**
+LLMs love QA formats because user prompts are usually questions.
+Transform prose into rigid QA objects using standard semantic data blocks.
+
+```html
+<dl>
+  <dt><h3>What is the data retention limit for the Free Tier?</h3></dt>
+  <dd><p>The Free Tier limits data retention to precisely 14 rolling days.</p></dd>
+
+  <dt><h3>Does the platform support HIPAA compliance?</h3></dt>
+  <dd><p>Yes, Enterprise Tiers support full BAA HIPAA compliance parameters.</p></dd>
+</dl>
+```
+
+---
+
+## 4. API Docs Readability (The Primary AI Target)
+
+When an AI tries to write code using your product, it scrapes your API documentation.
+
+1. **Eliminate Image-Based Architecture:** Highlighting your architecture solely in a PNG graphic is invisible. Describe the system architecture using text or standard Mermaid.js code blocks.
+2. **Provide Concrete Copy-Paste Examples:** The easiest way to get an LLM to use your platform correctly is to ensure your docs contain perfectly working `curl`, `TypeScript`, and `Python` code snippets with zero external dependencies.
+
+---
+
+---
 
+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.
+
+---
+
+**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.
 
 ---
 
@@ -175,6 +174,7 @@ AI coding assistants often fall into specific bad habits when dealing with this
 ### ✅ 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?
@@ -185,5 +185,6 @@ 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.