RAG Demo -- Vector Search Support Agent

Acme Corp support agent with vector RAG retrieval. Demonstrates the full vector path: Document -> Chunker -> Embedder -> VectorStore -> VectorRetriever -> createVectorRetrievalTool -> Agent -> Runtime -> Hono.

What It Shows

• Ingesting markdown documents into a vector store (chunk, embed, upsert)
• Retrieval via cosine similarity with RagPipeline
• Agentic metadata filters (LLM constructs filters at query time)
• Swappable vector stores: InMemory, PostgreSQL (pgvector), Redis (Redis Search)
• Session persistence across turns
• Hono server with SSE and WebSocket endpoints

Setup

bun install

Copy .env.example to .env and add your OpenAI API key.

Run

# In-memory (default, no external deps)
bun run dev          # HTTP server on :3334
bun run cli          # Interactive CLI

# PostgreSQL + pgvector
DATABASE_URL=postgresql://localhost:5432/ariaflow_rag bun run dev:pg

# Redis + Redis Search
REDIS_URL=redis://localhost:6379 bun run dev:redis

Vector Store Setup

PostgreSQL

createdb ariaflow_rag
psql -d ariaflow_rag -c "CREATE EXTENSION IF NOT EXISTS vector"

Redis

Requires Redis Stack (includes Redis Search module):

brew install redis-stack-server
redis-stack-server --daemonize yes

File Structure

rag-demo/
  rag.ts        -- Pipeline setup: Embedder + VectorStore + Chunker + RagPipeline
  agent.ts      -- Agent with createVectorRetrievalTool (agentic filters enabled)
  server.ts     -- Hono server (SSE, WebSocket, health)
  cli.ts        -- Interactive CLI with inline search/result display
  knowledge/
    policies.md -- Refund, shipping, privacy, warranty, billing policies
    products.md -- Plans (Starter/Pro/Enterprise), Widget X100, Cloud Backup

Endpoints

| Endpoint | Method | Description | |----------|--------|-------------| | /health | GET | Health check | | /api/chat/sse | POST | SSE streaming | | /api/chat/stream | POST | Plain text streaming | | /api/chat | POST | Full response | | /ws/:sessionId | WS | WebSocket streaming |

Environment Variables

| Variable | Required | Default | Description | |----------|----------|---------|-------------| | OPENAI_API_KEY | Yes | -- | OpenAI API key | | VECTOR_STORE | No | memory | memory, pg, or redis | | DATABASE_URL | When pg | -- | PostgreSQL connection string | | REDIS_URL | When redis | redis://localhost:6379 | Redis connection URL | | PORT | No | 3334 | Server port |