MCP ASDK Studio v0.9-4 (Public Beta) 🚀

"Your High-Performance, Free Alternative to Commercial AI Desktops"

Professional Setup Interface: Connect local scripts, configure AI profiles, or remote SSE servers with ease.

Live Workspace Evidence

MCP ASDK Studio provides a practical, desktop-first environment for AI tool orchestration.

It connects to live MCP servers, exposes visible tool inventories, and generates fact-grounded outputs from multiple sources including market data, SEC filings, and news summaries.

Typical workflow:

1. Connect MCP servers
2. Inspect available tools
3. Ask a ticker or domain question
4. Review the combined analysis
5. Expand raw tool results for verification

Live MCP server connections and tool inventory

Structured multi-tool analysis inside the workspace

Expandable raw outputs for inspection and verification

MCP ASDK Studio is a 100% Free, Open-Source AI development workspace. (v0.9-4 Public Beta) Actual Product UI - Experience the premium 'Lim Chat PRO' engine for free.

---

💎 Why choose ASDK Studio?

• Commercial-Grade Engine: Migrated from the verified Lim Chat PRO architecture.
• True Alternative: Full support for MCP servers and custom AI providers, rivaling commercial chatbot desktops.
• AI-Powered Customization: Built with a clean Python/JS structure. Use AI to fix, modify, and expand the studio to fit your unique needs.
• 100% Privacy: No tracking. Your keys and data stay on your local machine.

🚀 Quick Start

1. Clone: git clone https://github.com/lim-asdk/asdk_09-mcp_asdk_studio_v0_9-4.git
2. Setup: Copy user_data/profiles/profile.sample.json to default.json and add your API key.

---

🏗️ Architecture: L1-L4 Vertical Alignment

[L4] WILL (Intelligence)  : Personas, Prompts, Logic Flow
            ↑
[L3] BRIDGE (Orchestrator): ProBridgeAPI, ExpertRunner, Tool Router
            ↑
[L2] LOGIC (Processing)   : Data Filtering, Auth Bridge, Reasoner
            ↑
[L1] PHYSICAL (Infra)     : PathManager, user_data, keys, .env

⚡ V5 Bootstrap Guide

Get the system running in 4 easy steps:

1. Environment: Create a .env file (optional) to override DATA_ROOT.
2. Dependencies: pip install -r requirements.txt
3. Authentication: Place JSON keys in the keys/ directory and set GOOGLE_APPLICATION_CREDENTIALS if needed.
4. Diagnostics & Launch:
   • Run python check_health.py to verify system integrity.
   • Run python main.py to launch the studio.

📂 Project Structure

• main.py: Desktop Launcher & Entry Point.
• check_health.py: System Integrity Diagnostic Tool.
• lim_chat_pro/: Core Vertical AI Engine & UI Assets.
• user_data/: Local private data (Profiles, History, MCP configs). (Git Excluded)
• keys/: Secure authentication keys storage. (Git Excluded)
• docs/: Multi-language documentation and reports.

---

📖 Documentation

• Quick Start Guide (EN) / 빠른 시작 (KO)
• English Setup Guide
• Auth Setup Guide
• Public Release & AI Optimization Guide
• Development Reports

© 2026 lim_hwa_chan. Released for the community.

?? Future Blueprint: V6 Matrix System & APLC Architecture

While this v0.9-4 Public Beta provides a robust desktop-first workspace, it is fundamentally the stepping stone towards the V6 Matrix System?an infinitely scalable, multi-surface intelligence architecture. Future updates will transition this single-instance studio into a decentralized APLC (AI Program Logic Controller) network.

Here is the architectural roadmap for upcoming iterations:

1. Room-Based Infinite Expansion

We are shifting from standard software logic to hardware-like intelligence orchestration.

• Zero-Code Cloning: Scaling the system will be as simple as copying the room_01 directory to room_02, room_03, etc.
• Component Isolation: Each "Room" acts as an independent memory module (RAM), capable of housing different models (GPT-4, Claude, Gemini) or distinct API keys. This guarantees maximum stability, akin to plugging parallel hardware components from different manufacturers into a single motherboard.

2. Cool Fail-over & The MMU Router

The routing mechanism will no longer process complex AI states. It will operate purely as a high-speed delivery protocol.

• Blind Delivery Protocol: If a Room fails to respond, the Router instantly drops the payload into the adjacent Room's inbox. The system never halts.
• MMU-Style Health Sweeps: On boot, the system conducts a comprehensive sweep of all Rooms, assigning strict [Healthy] or [Bad] labels. The Router functions like a Memory Management Unit (MMU), delivering data only to verified [Healthy] addresses.

3. Multi-Surface Architecture

The system will be physically divided into isolated operational planes:

• User Surface: A hyper-simplified UI for final outputs.
• Operator Surface: For system monitoring, prompt orchestration, and overriding AI actions.
• Developer Surface: The core logic and routing manipulation plane.

Note to Contributors: All future pull requests and feature proposals must align with the Pointer-Only principle (passing coordinates, not duplicating data) and the physical isolation of the L1-L4 intelligence layers.