This repository provides a comprehensive Javascript-based AI stack for rapid prototyping of weekend projects. It integrates essential components like LLM orchestration, vector databases, authentication, and deployment configurations, targeting developers looking to quickly build and deploy AI-powered applications.

How It Works

The stack leverages Langchain.js for LLM orchestration, enabling seamless integration with various AI models and vector stores. It supports OpenAI for text generation and Replicate for image models. Data is stored and queried using either Pinecone or Supabase's pgvector, with embeddings generated via a provided script. Authentication is handled by Clerk, and deployment is streamlined for platforms like Fly.io.

Quick Start & Requirements

• Install dependencies: npm install
• Fill secrets in .env.local (Clerk, OpenAI, Replicate, Pinecone/Supabase, Arcjet API keys).
• Generate embeddings: npm run generate-embeddings-pinecone or npm run generate-embeddings-supabase.
• Run locally: npm run dev
• Deployment: fly launch then fly deploy --ha=false.
• Prerequisites: Node.js, npm, API keys for services used.
• Official Docs: Langchain.js

Highlighted Details

• Full-stack Javascript AI boilerplate.
• Supports multiple vector stores (Pinecone, Supabase pgvector).
• Integrates OpenAI and Replicate models.
• Includes authentication (Clerk) and security (Arcjet).
• One-click deployment to Fly.io.

Maintenance & Community

• Community Discord available: AI Stack Devs.
• Contributions are welcomed via pull requests.

Licensing & Compatibility

• License: Not explicitly stated in the README.
• Compatibility: Designed for Javascript environments (Node.js, Next.js).

Limitations & Caveats

The README does not specify the project's license, which is a critical detail for commercial use or closed-source linking. The setup requires obtaining multiple third-party API keys, which may incur costs.