NavyDevilDoc/rag-advisor

RAG Architecture Advisor is a small full‑stack tool that guides developers/architects through a 12‑question wizard to recommend an optimal retrieval‑augmented generation (RAG) architecture (Standard Vector, Graph, or Agentic). It also provides deployment‑paradigm guidance and AI‑generated reasoning using the Anthropic API. Who uses it: engineers or solution architects evaluating RAG approaches for a specific use case. Key features: - 12‑question wizard that outputs architecture recommendation and rationale (Anthropic) - Deployment guidance and a single‑service deployment pattern - Frontend: Vite + React 18; Backend: FastAPI + Uvicorn + Anthropic Python SDK - Local dev scripts, Vite proxy for /api/*, and Railway deployment via railway.json - Docs and design notes in CLAUDE.md and build status in PROGRESS.md Prereqs: Node 18+, Python 3.11+, uv package manager, and an Anthropic API key.

AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.