Merge pull request #11 from rodrigooler/codex/vinext-apps-migration

migrate apps to vinext and clean nextjs artifacts

Author: rodrigooler

.github/workflows/q2-validate.yml

@@ -77,7 +77,7 @@ jobs:
       - name: Setup Node
         uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020
         with:
-          node-version: "20"
+          node-version: "22"
           cache: "npm"
           cache-dependency-path: apps/explorer/package-lock.json
 

AGENTS.md

@@ -132,6 +132,15 @@ This is the only valid definition. Do not treat any intermediate state as final.
 
 ---
 
+## Commit Convention
+
+- Use Conventional Commits for every commit message.
+- Preferred prefixes: `feat:`, `fix:`, `refactor:`, `chore:`, `docs:`, `test:`.
+- Add scope when it increases clarity, for example: `feat(site): ...` or `fix(programs): ...`.
+- Keep the subject line imperative and specific. Avoid vague messages like `update stuff`, `changes`, or `wip`.
+
+---
+
 ## Coding Standards
 
 ### Rust (runtime, programs, SDK)

CLAUDE.md

@@ -26,6 +26,13 @@ Active work:
 
 **Do not** ask Claude to implement Testnet/Mainnet-only scope (GPU proving, full 6 challenge categories, TGE tokenomics) yet.
 
+## Commit Convention
+
+- Use Conventional Commits for all commits.
+- Default prefixes: `feat:`, `fix:`, `refactor:`, `chore:`, `docs:`, `test:`.
+- Use scoped messages when helpful, for example `feat(site): ...`.
+- Do not use vague commit subjects like `update`, `misc`, or `wip`.
+
 ---
 
 ## Decision Log (Why We Made Key Choices)
@@ -111,7 +118,7 @@ Current phase: Phase 1 (Q2 2026) — Devnet implementation and validation.
 Next milestone: Testnet preparation (Q3/Q4 2026).
 
 Key files:
-- Whitepaper: docs/whitepaper/RAXION_Whitepaper_v1.0_EN.md
+- Whitepaper: docs/whitepaper/RAXION_Whitepaper_v0.4_EN.md
 - Core formulas: Chapter 3 of whitepaper
 - Architecture: AGENTS.md
 

MEMORY.md

@@ -17,6 +17,12 @@ GitHub: https://github.com/raxion-network/raxion
 License: BUSL 1.1 → MIT 2030-02-20
 ```
 
+### Working Convention (2026-02-28)
+
+- Commit messages now follow Conventional Commits.
+- Approved prefixes: `feat:`, `fix:`, `refactor:`, `chore:`, `docs:`, `test:`.
+- Prefer scoped messages when useful, e.g. `feat(site): ...`.
+
 ### Q2 Execution Notes (2026-02-25)
 
 - Dockerized deploy path added for Anchor/Solana:
@@ -239,7 +245,7 @@ If challenged, finality is resolved only by challenge verification pass.
 
 ## Parameter History
 
-> Track all parameter changes with dates and rationale. Initialized at whitepaper v1.0 values.
+> Track all parameter changes with dates and rationale. Initialized at whitepaper v0.4 values.
 
 ### Active Parameters (as of 2026-02-22)
 

WHITEPAPER.md

@@ -171,7 +171,7 @@ Zero-Knowledge Proofs for Machine Learning (zk-ML) were impractical until 2023
 
 Three recent developments changed this equation:
 
-**Next-generation ZK frameworks**: Jolt [11] and RISC Zero v1.0 [12] drastically reduced proof generation overhead for generic computation, including ML inference. What once took hours now takes minutes. What took minutes is converging to seconds with ongoing software optimizations.
+**Next-generation ZK frameworks**: Jolt [11] and RISC Zero v0.4 [12] drastically reduced proof generation overhead for generic computation, including ML inference. What once took hours now takes minutes. What took minutes is converging to seconds with ongoing software optimizations.
 
 **Dedicated ZK hardware**: The first generation of ASICs and FPGAs optimized for ZK proof generation is beginning to reach market. Companies like Ingonyama and others are developing hardware that can accelerate proof generation by 10–100x relative to current CPU/GPU implementations.
 
@@ -776,7 +776,7 @@ fn respond(query: Query, memory: &AgentMemory) -> Response {
 }
 ```
 
-**RaxLang Roadmap**: Rust-only on Devnet → RaxLang v0.1 on Testnet → RaxLang v1.0 on Mainnet.
+**RaxLang Roadmap**: Rust-only on Devnet → RaxLang v0.1 on Testnet → RaxLang v0.4 on Mainnet.
 
 ### 2.6.3 zk-ML Integration: Jolt and RISC Zero
 
@@ -1268,14 +1268,14 @@ royalty_A = gas_total(inference_B) × cross_agent_usage_fraction × 0.10
 
 **Deliverables:**
 - [x] Whitepaper v0.1, v0.2, v0.3
-- [ ] **Whitepaper v1.0** (this document)
+- [ ] **Whitepaper v0.4** (this document)
 - [ ] Agave fork technical specification (Neural SVM Runtime v0.1)
 - [ ] Local PoC: 2 architectures processing queries without ZK proof
 - [ ] Initial RISC Zero integration for simple model proof-of-execution
 
 **Success Criteria:**
 ```
-✓ Whitepaper v1.0 published
+✓ Whitepaper v0.4 published
 ✓ PoC demonstrates semantic convergence between 2 architectures in >70% of queries
 ✓ Execution proof generated for simple model
 ✓ At least 3 external developers contributing to repository
@@ -1382,7 +1382,7 @@ $RAX gains real economic value — activating true protocol incentives. This is
 
 ```
 2026 Q1  ──── Phase 0: Genesis
-               Whitepaper v1.0, local PoC, technical specification
+               Whitepaper v0.4, local PoC, technical specification
 
 2026 Q2  ──── Phase 1 start: Devnet
                Neural SVM Runtime v0.1, PoIQ v0.1, 10 agents
@@ -1400,7 +1400,7 @@ $RAX gains real economic value — activating true protocol incentives. This is
                $RAX listed on exchanges, on-chain governance active
 
 2027 Q4  ──── Mainnet v1 stabilization
-               RaxLang v1.0, 50+ Smart Agents with RS > 0.80
+               RaxLang v0.4, 50+ Smart Agents with RS > 0.80
 
 2028     ──── Phase 4 start: Mainnet v2
                Sub-second latency, Agent Coordination Networks
@@ -1481,7 +1481,7 @@ For researchers who want to contest the PoIQ formalization: the technical Yellow
 
 [11] Arun, S. et al. "Jolt: SNARKs for Virtual Machines via Lookups." a16z crypto research (2024). jolt.a16zcrypto.com
 
-[12] RISC Zero. "RISC Zero v1.0 — Boundless: A Universal ZK Coprocessor." (2024). risczero.com/news/boundless
+[12] RISC Zero. "RISC Zero v0.4 — Boundless: A Universal ZK Coprocessor." (2024). risczero.com/news/boundless
 
 [13] EZKL. "Machine Learning Proofs — Documentation." (2025). ezkl.xyz/docs
 
@@ -1531,7 +1531,7 @@ For researchers who want to contest the PoIQ formalization: the technical Yellow
 
 ---
 
-*RAXION Whitepaper v1.0 — February 2026*
+*RAXION Whitepaper v0.4 — February 2026*
 *Document License: CC BY 4.0*
 *Code License: BUSL 1.1 → MIT on 02/20/2030*
 

apps/explorer/app/globals.css

@@ -21,7 +21,7 @@ body {
     radial-gradient(circle at 12% 18%, #1c4267 0%, transparent 45%),
     radial-gradient(circle at 84% 8%, #3d2f59 0%, transparent 38%),
     linear-gradient(180deg, #06101a 0%, #08141f 100%);
-  font-family: var(--font-display), sans-serif;
+  font-family: "Space Grotesk", "Inter", "Segoe UI", sans-serif;
 }
 
 a {
@@ -62,7 +62,7 @@ a:focus-visible {
 .small {
   font-size: 12px;
   color: var(--muted);
-  font-family: var(--font-mono), monospace;
+  font-family: "IBM Plex Mono", "JetBrains Mono", ui-monospace, monospace;
 }
 
 .score-track {
@@ -82,7 +82,7 @@ a:focus-visible {
 .table {
   width: 100%;
   border-collapse: collapse;
-  font-family: var(--font-mono), monospace;
+  font-family: "IBM Plex Mono", "JetBrains Mono", ui-monospace, monospace;
   font-size: 12px;
 }
 

apps/explorer/app/layout.tsx

@@ -1,19 +1,14 @@
-import type { Metadata } from "next";
-import { IBM_Plex_Mono, Space_Grotesk } from "next/font/google";
 import "./globals.css";
 
-const display = Space_Grotesk({ subsets: ["latin"], variable: "--font-display" });
-const mono = IBM_Plex_Mono({ subsets: ["latin"], weight: ["400", "500"], variable: "--font-mono" });
-
-export const metadata: Metadata = {
+export const metadata = {
   title: "RAXION Devnet Explorer",
   description: "Live PoIQ convergence feed from Solana devnet",
 };
 
 export default function RootLayout({ children }: Readonly<{ children: React.ReactNode }>) {
   return (
     <html lang="en">
-      <body className={`${display.variable} ${mono.variable}`}>{children}</body>
+      <body>{children}</body>
     </html>
   );
 }