---
name: al-701-build-a-retail-shopping-and-service-superagent-one-orches
description: "Salesforce's Nov 3, 2026 customer story describes how Crocs, with Rosetree Solutions and Salesforce forward-deployed engineers, shipped Rivet—an Agentforce retail superagent—in five months. A single orchestrator coordinates specialist shopper and service agents across five launch use cases: product discovery, FAQs, order tracking, returns and conversational checkout, with real-time order/return data from IBM Sterling. The same service actions were reused to launch Agentforce Voice in seven weeks. Federated Data 360 spaces keep Crocs and HEYDUDE brand data separate while supporting shared reporting. Use when the user wants help with retail multi-agent superagent architecture (ecommerce) for a small business."
---

# Build a retail shopping-and-service superagent: one orchestrator routes customers to specialist shopper and service agents for discovery, sizing, orders, returns and checkout

Salesforce's Nov 3, 2026 customer story describes how Crocs, with Rosetree Solutions and Salesforce forward-deployed engineers, shipped Rivet—an Agentforce retail superagent—in five months. A single orchestrator coordinates specialist shopper and service agents across five launch use cases: product discovery, FAQs, order tracking, returns and conversational checkout, with real-time order/return data from IBM Sterling. The same service actions were reused to launch Agentforce Voice in seven weeks. Federated Data 360 spaces keep Crocs and HEYDUDE brand data separate while supporting shared reporting.

## Inputs to collect first
Ask the user for anything below that they haven't already given you. Don't guess missing facts.

- Brand
- Web chat / voice
- Criteria

## Instructions

Do the task below with the user's inputs in place of each [bracket].

Design a retail superagent for [brand] like Crocs Rivet. Channels: [web chat / voice]. First release use cases only: product discovery, FAQs, order tracking, returns, conversational checkout. Architecture: one Orchestrator + Shopper agent + Service agent; list tools each owns (catalog, sizing/fit KB, OMS APIs for status/returns). Guardrails: never invent inventory or refund eligibility; hand off to human when [criteria]. 5-month build plan with a 7-week voice reuse path. Analytics: autonomous resolution rate and FAQ accuracy. Do not claim live production metrics we have not measured.

## Safety checks (share these with the user)

- Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes.
- Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings.
- Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.
- Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed.

## About this skill

- Works with: Salesforce Agentforce, Service Cloud, Commerce Cloud, Data 360, IBM Sterling OMS
- Based on this real-world example: https://www.salesforce.com/uk/customer-stories/crocs/building-ai-agents/
- Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-701-build-a-retail-shopping-and-service-superagent-one-orchestra/
- To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/