---

🏛️ linacre.site — Enterprise Developer Portal & Autonomous AI Ecosystem

linacre.site is an offline-first developer portal, multi-model AI orchestration lab, and personal software engine engineered by David Linacre. Built with React 19, TypeScript, Vite 6, and Express 4, deployed globally across Vercel Edge with zero-downtime static prerendering across all 24 routes.

CI status note: tsc typecheck + full build verified locally (see PR #4). GitHub Actions is currently paused by an account-level billing lock (free plan, no minutes consumed) — jobs show as failed without ever starting a runner. Workflow disabled until billing is cleared; one click re-enables it (repo → Actions → CI → Enable).

---

⚡ System Architecture

flowchart TD
    subgraph Client ["Client Browser (PWA Shell)"]
        ReactSPA["React 19 SPA Container"]
        SWCache["Service Worker Cache (sw.js)"]
        StateStore["LocalStorage & IndexedDB State"]
    end

    subgraph Edge ["Vercel Edge & Content Delivery"]
        EdgeCDN["Vercel Global CDN"]
        CSP["Content Security Policy Engine"]
        StaticPre["Prerender Snapshot Engine (24 Routes)"]
    end

    subgraph Backend ["Serverless API Backend (api/server.ts)"]
        ExpressApp["Express API Engine"]
        AuthMiddleware["SHA-256 Timing-Safe Auth Guard"]
        RateLimiter["IP Sliding Window Rate Limiter"]
    end

    subgraph LLM ["AI Engine Integration"]
        Gemini["Google Gemini 3.6 Flash API"]
        OpenAI["OpenAI GPT-4o Fallback"]
        Claude["Anthropic Claude 3.5 Sonnet"]
        LocalOllama["Local Docker Ollama Instance"]
    end

    ReactSPA --> EdgeCDN
    EdgeCDN --> StaticPre
    ReactSPA --> ExpressApp
    ExpressApp --> AuthMiddleware
    ExpressApp --> RateLimiter
    ExpressApp --> Gemini
    ExpressApp --> OpenAI
    ExpressApp --> Claude
    ExpressApp --> LocalOllama

Loading

---

🌟 Key Application Hubs

• ⚡ AI Lab (/lab): Unified multi-provider model sandbox with active streaming proxies for Gemini 3.6 Flash, OpenAI, Claude, and local Docker Ollama nodes. Features pre-tuned Quick-Start Engineering Templates (Code Audit, PostgreSQL Schema Designer, Regex Builder, OpenAPI 3.0 Spec Generator).
• ✨ Identity Studio (/identity): Real-time AI prompt emblem generator powered by Gemini 3.6 Flash with dynamic brand palette matching and SVG/PNG vector exports.
• 🧰 Developer Playground (/playground): Client-side engineering utilities including JSON-to-TypeScript Generator with JSDoc headers, 5-field Cron Expression Explainer & Schedule Builder, JWT Inspector, RegEx Evaluator, and WebAssembly compiler testbed.
• 🤖 Agents Hub (/agents): Autonomous agent workflow visualizer, step execution simulator, and one-click JSON blueprint exporter for AI agent playbooks.
• 📅 Discovery Scheduling (/book): Dedicated client route for 15-minute discovery calls and 45-minute architectural review appointments.
• 📊 Telemetry Dashboard (/dashboard): Real-time service mesh monitoring with accessible text data tables, radar stats, and KeePassXC secret key state verification.

---

🔒 Security & Governance

• Zero Hardcoded Secrets: Centralized secrets pipeline managed via linacre.py CLI and environment variables (.env).
• Timing-Safe Authentication: Session authentication uses HMAC-SHA256 signatures with constant-time equality comparisons (crypto.timingSafeEqual) to mitigate side-channel timing attacks.
• Strict CSP Headers: Hardened Content-Security-Policy rules in vercel.json enforcing strict script, style, frame, and font origin restrictions.
• Bounded Rate Limiting: Sliding-window IP rate limiters on all public API endpoints to prevent brute-force attacks and resource exhaustion.

---

🛠️ Local Development & Build Pipeline

Prerequisites

• Node.js >= 20.x
• npm >= 10.x
• Python 3.11+ (optional for orchestration via linacre.py)

Development Commands

# Install dependencies
npm install

# Launch development environment (Vite middleware + Express API on port 3000)
npm run dev

# Run TypeScript typechecks & linter
npm run lint

# Build production bundle & prerender static HTML snapshots (24 routes)
npm run build

# Preview production build locally
npm run start

Deployment Command

# Automated build and production deployment via Python CLI
python linacre.py build
vercel deploy --prod --yes --project linacre-site-repo --force

---

👨‍💻 About the Author

David Linacre
Principal Infrastructure Engineer, DevOps Architect, and Software Systems Developer

• 🌐 Live Portal: www.linacre.site
• 🐙 GitHub Org: @LIN4CRE
• 👤 GitHub Personal: @DLinacre
• ☕ Sponsor Craft: paypal.me/DLinacre16

---

📄 License

This repository is licensed under the MIT License.