Multi-Agent PPC Management with Claude Code for Amazon Sellers

What they did

A Reddit seller describes running a multi-agent setup on their own account for months, with separate agents for bidding, inventory and strategy and rules-based conflict priority between them. Data comes in through SP-API and the Ads API, with search-term and campaign data fed to the agents as context.

How it went

They report about a week of setup ('3 hrs per/day of sleep and a month of Claude Max') and had to work around SQP and search term report API timeouts. Their worst mistakes came from an agent reacting to one bad day of data and from two agents making conflicting moves. Manual tweaks made while the system was running 'tanked my entire system' and cost margin that month, which they say they're making back.

We only report what the source says. Results are self-reported unless noted.

Worth knowing

Their advice: start in suggestion mode, keep the approve/dismiss history to tune the rules, and automate negative-keyword harvesting before bid changes. Another commenter in the thread warns that beyond PPC you need a proper data warehouse.

Copy-paste prompt

You are the strategy agent in my Amazon PPC system. You do NOT execute changes. Each day:
1. Read the bidding agent's proposed bid changes and the inventory agent's stock alerts (pasted below).
2. Resolve conflicts with this priority: inventory risk > profitability > growth. Example: never approve a bid increase on an ASIN with < 14 days of stock.
3. Ignore any recommendation driven by a single day of data unless it's confirmed by the 7-day trend.
4. Output an approval queue for me: change, agent, reason, conflict resolved (Y/N).
Bidding agent proposals: [paste]
Inventory agent alerts: [paste]

The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].