🧠 Cortex Gate

Extensible AI Platform — Modular gateway for controlling AI token flows via plugins.

Cortex Gate is an extensible platform for routing, governing, and optimizing AI LLM requests. It provides a pluggable architecture where functionality is delivered through extensions — self-contained modules that register UI pages and backend routes.

Current status: Early v0.2. The platform shell and extension system are functional. Individual extensions (routing, governance, benchmarking) are under active development.

---

✨ Current Features

Feature	Status	Description
Extension System	✅ Done	Registry with lazy-loaded pages, toggle enable/disable, localStorage persistence
Prompt Router UI	✅ Done	Visual equalizer with 8 dimensions + economy slider
Server Control	✅ Done	Tauri app manages backend lifecycle (start/stop/status)
Backend API	✅ Done	OpenAI-compatible /v1/chat/completions endpoint
SQLite Database	✅ Done	User, API key, and budget storage
Provider Config	✅ Done	Multi-provider configuration (OpenAI, Anthropic, OpenRouter)
Embedding Classifier	🟡 Stub	ONNX int8 embeddings — keyword heuristic fallback for now
Benchmark Engine	🟡 Stub	Real model benchmarking — test framework exists, full integration pending
Cost Governance	🟡 Stub	Token budgets and alerts — basic structure in place
Streaming Proxy	🟡 Stub	SSE passthrough + Anthropic→OpenAI translation — wired but untested with real providers

---

🗺️ Roadmap

v0.2 (current) — Extensible Foundation

• Extension registry (frontend: JS, lazy imports, toggle enable/disable)
• Prompt Router extension (equalizer UI)
• Server control from Tauri (start/stop/status)
• Backend serves frontend static files (single port :18801)
• Wire classifier → actual ONNX embeddings
• Wire streaming → real provider calls
• Wire governance → token tracking + budget enforcement

v0.3 — Functional Routing

• Complete embedding classifier (ONNX int8, dimension scoring)
• Real provider proxy (streaming SSE with keepalive)
• Token usage tracking + cost calculation
• Dashboard extension (real-time stats, charts)
• Rate limiting middleware (tower-governor)

v0.4 — Production Hardening

• Circuit breaker per provider (tower-resilience)
• Retry with exponential backoff
• OpenAPI docs (utoipa-axum)
• Stress testing + performance benchmarks
• Prometheus metrics

v0.5 — Extension Ecosystem

• Analytics extension
• Audit log extension
• A/B testing extension
• MCP / JSON-RPC protocol support

---

🏗️ Architecture

┌──────────────────────────────────────────────────┐
│                   Client                          │
│  (OpenAI-compatible SDK / HTTP)                   │
└──────────┬───────────────────────────────────────┘
           │ POST /v1/chat/completions
           ▼
┌──────────────────────────────────────────────────┐
│              Gateway (Rust / Axum)                │
│  :18801/api/*  +  :18801/  (static frontend)     │
│                                                   │
│  ┌──────────┐  ┌──────────┐  ┌────────────────┐  │
│  │ Auth     │  │ Routes   │  │ Extensions     │  │
│  │ Middleware│  │ Handlers │  │ (plugin loader)│  │
│  └──────────┘  └──────────┘  └────────────────┘  │
│                                                   │
│  ┌──────────┐  ┌──────────┐  ┌────────────────┐  │
│  │ProxyEngine│  │Classifier│  │ Governance     │  │
│  │(reqwest) │  │(ONNX)    │  │ (SQLite)       │  │
│  └──────────┘  └──────────┘  └────────────────┘  │
└──────────────────────────────────────────────────┘
           │
           ▼
┌──────────┴──────────┐     ┌──────────────────────┐
│   LLM Providers     │     │   Tauri Desktop App  │
│  OpenAI / Anthropic │     │  (Electron-like,     │
│  OpenRouter / Custom│     │   embeds frontend)   │
└─────────────────────┘     └──────────────────────┘

---

🚀 Quick Start

# 1. Build and run the backend
cargo build --release
./target/release/cortex-gate

# 2. Open in browser
# http://127.0.0.1:18801

# Or use the Tauri desktop app (Linux with display):
cd frontend/src-tauri && cargo build --release
DISPLAY=:0 ./target/release/cortex-gate-tauri

Tauri Desktop Controls

1. Double-click the Cortex Gate icon
2. Click Start Server in the sidebar
3. Status dot turns green → Server Online
4. Click Open Web UI → browser opens http://127.0.0.1:18801
5. Click Stop Server to shut down

---

🧩 Extension System

Extensions are self-contained modules that register UI pages and optional backend routes. Currently available:

Extension	Status	Description
Prompt Router	🟡 Partial	Equalizer UI works; backend routing not yet wired to real providers

To create a new extension:

// frontend/src/extensions/my-ext/manifest.js
export const manifest = {
  id: "my-ext",
  name: "My Extension",
  description: "Does something cool",
  version: "0.1.0",
  pages: [{
    name: "my-ext",
    label: "My Ext",
    icon: "puzzle",
    load: () => import("./page.js"),
  }],
};

Then register it in frontend/src/main.js.

---

📖 Documentation

• Project Origins — Historical context
• Best Practices — Code standards and patterns
• Code Suggestions — Pending improvements

---

⚖️ License

PolyForm Noncommercial License 1.0.0. See LICENSE.

For commercial use, contact the author: Luis Daniel Dos Santos