Multi-Agent System

AI agents that actually
run your store

Four specialized agents with real database access. Ask about revenue, generate product copy, check inventory, plan campaigns — they query actual data and give you real answers.

The Agents

Each one has its own system prompt, typed tools, and domain knowledge. The orchestrator picks who handles what.

Analytics Agent

Revenue trends, top products, customer segments, traffic sources, and anomaly detection.

query_revenuetop_productsrevenue_over_timedetect_anomaliescustomer_segmentstraffic_sourcessales_by_category

Content Agent

Product descriptions, SEO optimization, pricing analysis, and bulk listing improvements.

generate_descriptionoptimize_seobulk_improve_listingssuggest_pricing

Inventory Agent

Stock monitoring, restock recommendations, demand forecasting, and inventory updates.

check_stock_levelsrestock_recommendationsupdate_stockforecast_demand

Marketing Agent

Campaign planning, email copywriting, social media content, and discount strategies.

generate_campaignwrite_email_copydiscount_strategysocial_media_posts

Architecture

How the pieces fit together.

Supervisor Orchestrator

A central agent routes requests to specialized sub-agents using a flat tool map — no nested LLM calls.

Real Database Queries

Every tool queries a live Postgres database with 1,200+ orders, 200 customers, and 15k analytics events.

Multi-Step Reasoning

Complex requests are broken into workflows — the orchestrator plans, then executes each step sequentially.

Streaming Tool Calls

Watch agents think in real-time — tool calls stream with animated step indicators and rich result cards.

Stack

Serverless-first. Neon for zero cold-start DB, Vercel AI SDK for streaming, server components where it makes sense.

Next.js 15TypeScriptVercel AI SDK v6Claude SonnetDrizzle ORMNeon PostgresTailwind CSS v4shadcn/uiFramer MotionRechartsZod

Try it out

Dashboard is open to browse. Chat needs an access code.

StorePilot
Access-code protectedDeployed on Vercel