NodeForge AI

NodeForge AI is a production-ready Python CLI tool designed to scaffold, generate, and manage Node.js backend code using local LLMs (Ollama), RAG (ChromaDB), and tool execution.

The goal is to provide a highly modular, scalable, maintainable, and developer-friendly architecture that orchestrates multiple agents (planner, coder, debugger, etc.).

Architecture

• CLI (cli/): Built with Typer for an elegant, easy-to-use terminal interface.
• Core (core/): Contains the Orchestrator which manages the multi-agent task lifecycle.
• Agents (agents/): Individual agents (Planner, Coder, Reviewer) logic.
• LLM (llm/): Client connections to the local LLM via Ollama.
• RAG (rag/ & vector_db/): Vector abstraction base and ChromaDB implementation for localized code indexing.
• Tools (tools/): Tools for file manipulation, terminal execution, and search.

Setup Instructions

Prerequisites

• Python 3.9+
• Ollama running locally

1. Create and Activate Virtual Environment

Windows:

python -m venv .venv
.venv\Scripts\activate

Mac/Linux:

python -m venv .venv
source .venv/bin/activate

2. Install Dependencies

Install the required packages:

pip install -r requirements.txt

3. Environment Variables

Copy the example logic configuration:

cp .env.example .env

Adjust the .env file as needed to point to your local LLM model path.

Example CLI Usage

Generate a new feature (e.g., authentication routes):

node-agent generate auth

Debug and fix issues in the current scope:

node-agent fix

Analyze the codebase for improvements:

node-agent analyze

Index the project files into the ChromaDB vector database:

node-agent index