🧠 Nexus Hub (Local Intelligence Hub)

A unified Monolithic API Server for Ollama. The Nexus Hub provides the architecture for System Mapping, Task Farming, Local Consultation, and Tool Orchestration.

Installation • Usage • Architecture

---

🛑 The Problem

Running multiple local Ollama tools simultaneously (like a consultant CLI, a background orchestrator server, a separate SQLite database for codebase mapping, and an asynchronous task queue) creates immense fragmentation and resource contention.

💡 The Solution

The Nexus Hub is a Monolithic API Server that provides a unified control plane for your local intelligence stack. By relying on a single backend, Native Ollama Functions, and a Persistent ChromaDB store, you can achieve <10ms latency routing without system bloat.

---

⚙️ Architecture

flowchart TD
    A[Nexus Hub API] --> B{Ollama Ecosystem}
    
    B --> C[Orchestrator]
    C --> D[Native Tool Execution]
    
    B --> E[Task Farm]
    E --> F[Synchronous Threading Queue]
    
    B --> G[System Mapper]
    G --> H[Persistent ChromaDB]
    
    B --> I[Consultant]
    I --> J[Local Chat Generation]

Loading

---

🚀 Installation

Install directly via pip or use uv.

git clone https://github.com/axtontc/local-intelligence-hub
cd local-intelligence-hub
pip install .

Ensure you have a local Ollama instance running.

---

💻 Usage (CLI)

Start the monolithic server:

nexus serve --port 8080

Trigger specific subsystems from the CLI:

nexus map --dir ./my_project
nexus farm --queue ./tasks.json
nexus consult --prompt "How do I optimize this architecture?"

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.