Building a Multi-Agent eCommerce Operations Team (2026)

AWS lifecycle notice (June 30, 2026) — Amazon Bedrock Agents Classic is in maintenance for new customers after July 30, 2026. Net-new agents should use Bedrock AgentCore. Full matrix: lifecycle roundup.

On June 17, 2026, AgentCore Harness reached general availability (What’s New). Export to Strands when a supervisor actually earns hop caps. Do not export because a slide says “operations team.”

This post assumes you already failed the split test in the supervisor direction. It is how to staff optional specialists — Sales, Customer, Inventory, Product, Finance — and how they fail when coordination is a prompt. It is not the store-agents sample duplicated. It is not an anonymized client engagement. Roster rows below are worksheets, not a FactualMinds headcount KPI.

First-party signals we reuse (not eCommerce outcomes) — Gateway server-side tools cut median tool round-trip ~180 ms → ~95 ms on a B2B CRM assistant (12 tools, ~8k turns/day) — Gateway post. Platform TCO silhouette: support-style AgentCore at 50K sessions/mo ~$791/mo platform + model (decision guide). That floor is one support-shaped mix. Five specialists are not five times the value; they are more sessions, more Gateway invokes, more ways to double-notify. Model your mix on the AgentCore pricing calculator.

Reproduce this — Copy supervisor-specialist-team.md. Fill reads vs writes per specialist with your APIs. Stubs: ecommerce-agentcore-store-agents/. Series folder: ecommerce-ai-agents-series/.

Opinionated take: the supervisor asks a closed question and a tool allow-list. Specialists do not wander. Trade-off: you give up a “team of agents that just figure it out” demo. You buy a consolidated decision a human can approve.

FactualMinds builds agents that connect storefronts, OMS, and back-office tools — with humans still on money, inventory, and identity.

The pattern (business question → consolidate)

Business Question → Supervisor → Delegation → Specialists → Consolidated decision

The supervisor is not a CEO chatbot. It is a router with a hop budget. If the question is “What’s the status of ORD-1001?”, you do not fan out to Finance. You call Customer (or stay on the original WISMO harness).

Strands 1.0 after agentcore export harness gives Agents-as-Tools, Graph, Swarm, Workflow. Prefer Agents-as-Tools or Graph for this roster. Swarm is the wrong default for money. Strands does not replace Gateway, Identity, Cedar, or Runtime isolation — ship map.

Specialist roster (optional — do not staff all five)

None of those writes become Execute because the box is named Finance. Autonomy is still per action. HITL still owns over-cap refunds, live price, large POs, and account changes.

The store-agents sample uses a slightly different four (sales, order ops, support triage, inventory). That is a cloneable topology, not a contradiction. Map names to your queues. Do not run both rosters.

Staffing order if you must add specialists this year: Customer (reads that WISMO already proved) before Inventory (ATP honesty), before Product (catalog draft), before Sales (promo-set reads), before Finance (margin fields). Finance last is intentional. Margin tools without a named margin data contract are a hallucination surface. Product before Sales because a sales specialist that cannot read required attrs will pitch SKUs the passport would fail closed.

Do not hire a “swarm of peers” to skip that order. Agents-as-Tools keeps the hierarchy; Graph freezes the refund/PO subgraph; Swarm is still the wrong money default.

Supervisor rules

1. Closed question. “Can we approve a replacement ship for ORD-1001 under the $75 worksheet cap?” — not “help with this purchase.”
2. Tool allow-list per hop. Inventory does not receive createReturn. Customer does not receive updateInventory.
3. Hop cap. Two specialist hops per turn is the sample default. Ambiguous “help with my purchase” must clarify, not fan out to five.
4. Evidence from each. Consolidate with evidence_tool (or equivalent) from each specialist that ran. A vibe summary is not a decision.
5. Writes stay off the specialists until Cedar is ENFORCE on that one action and a HITL queue exists for the over-cap path.
6. Identity still on Gateway. Shopper JWT must DENY associate tools even if the supervisor invoked Inventory by mistake.

flowchart LR
  Q[Closed question] --> Sup[Supervisor]
  Sup -->|allow-list A| A[Specialist]
  Sup -->|allow-list B| B[Specialist]
  A -->|evidence| C[Consolidate]
  B -->|evidence| C
  C --> Rec[Recommend / Draft]
  Rec --> HITL[HITL if write]

Next.js / TypeScript / Tailwind is a typical HITL dashboard. It is not the runtime.

Coordination challenges (this is the actual work)

Gateway ~180 → ~95 ms is tool RTT on the CRM canary. Five specialists in series will still wait on OMS p95. Do not promise a faster shopper because you added Finance.

Platform cost: model the sum of sessions on the AgentCore pricing calculator. The published ~$791/mo at 50K sessions is a silhouette for a support-style mix, not a budget for an unbounded swarm.

How this sits on AWS

Stay on one Harness until the split test says supervisor. Then export:

# Sketch — export Harness to Strands. Pin @aws/agentcore to what you install.
# Python 3.12+ on Runtime after export; boto3 ≥ 1.38.0; supported AgentCore region.
npx @aws/agentcore export harness \
  --harness-arn "arn:aws:bedrock-agentcore:us-west-2:123456789012:harness/commerce-ops" \
  --output ./commerce-ops-strands

Wire specialists as Agents-as-Tools (hierarchical) or a Graph (fixed edges for refund / PO subgraphs). Skip Swarm for refunds, ATP, and pricing. Workflow (Strands) is for deterministic subgraphs you should have left in Flow.

There is no native Shopify connector. Each specialist’s tools are OpenAPI / MCP / Lambda you own — integration. Agents Classic is the wrong net-new path after 30 July 2026.

Observability: CloudTrail on bedrock-agentcore.amazonaws.com plus Gateway traces. If you cannot show which specialist ran, you do not have a team. You have a conversation.

What broke

What broke — A merchant (worksheet, not a named engagement) exported a clean WISMO harness and added Sales + Inventory + Finance in one sprint so “ops could be covered.” Supervisor prompt: “Do what’s best for the customer and the margin.” Detection: Inventory recommended a clearance markdown; Finance recommended holding list; Customer specialist sent a delay notice and a win-back using two templates; no hop cap; Policy LOG_ONLY. Finance saw a draft price write with no merchandiser. Fix: strip all writes; one closed question per turn; hop cap 2; conflict rows go to HITL; Graph for any future price-recommend path; Agents-as-Tools only. Lesson: a roster without coordination rules is five prompts sharing a blast radius.

The already-published dual-write on delivered cancel still applies if any specialist receives cancelOrder — store-agents.

What to Do This Week

1. Confirm you actually failed the split test toward a supervisor. If not, stop.
2. Copy supervisor-specialist-team.md. Check one specialist you might add this quarter — not five.
3. Write the closed question that specialist will answer. If you cannot, you do not need them.
4. Name read tools only. Browser off.
5. Set hop cap 2 in the export target before the first specialist invoke.
6. Stand up HITL before any specialist write — HITL.
7. Price the extra sessions on the AgentCore pricing calculator.
8. Run monday-checklist.md. Contact us if the first specialist is Finance.

What This Post Doesn’t Cover

• Whether you need multiple agents at all — post 57
• Cloneable supervisor stubs — store-agents
• CTO stack composition — post 12
• Memory retention policy — post 59
• Measured ops-hours saved from a named five-agent team — we are not inventing them

FAQ

When should you NOT staff all five eCommerce specialist agents?

Skip the full roster in week one. Add a specialist only when prompt, tools, and IAM cannot stay inside the first agent.

What could go wrong if the supervisor delegates an open-ended “handle this order”?

Every specialist fires. Conflicts get averaged. Double-notify. Delegate a closed question plus a tool allow-list.

When should you NOT let specialists write?

Default writes to none. Quantity, live price, PII, and PO send stay HITL. Execute is per action, not per job title.

What could go wrong if specialists share Memory as the customer file?

Stale facts, cross-shopper leak, a shadow CRM. Tools for facts. Memory is conversational state.

Should money-adjacent specialists use Swarm?

No. Agents-as-Tools or Graph. Swarm explores. Export still does not replace Cedar.

Is this the store-agents supervisor sample duplicated?

No. This is the operations roster and coordination rules. Link the sample.

Need a supervisor roster with hop caps before the first specialist write? Contact FactualMinds or see Amazon Bedrock.