Seller Intel — an Ounie app for AI agents

Six Amazon-seller intelligence tools on one time-series moat: niche.finder (ranked sub-niche opportunities), demand.track (units-per-month estimates from BSR history), review.mine (build-a-better-version specs from review gaps), launch.guard (6-hourly price/BSR/stockout alerts), arbitrage.map (cross-marketplace gaps), and listing.doctor (audits vs category top performers).

Pay-per-run Ounie credits: niche sweep 200, demand refresh 35/ASIN, review report 1,900, guard pass 40, arbitrage run 9,900, listing audit 25/ASIN — failed runs refund automatically
Web dashboard, REST agents API, and a native MCP server for Cursor, Claude, and ChatGPT
Recurring monitors pause when the wallet runs short and resume on top-up

For agents

Works with Claude, ChatGPT, and any MCP client. One server, every app: paste the gateway config and the agent can discover and run Seller Intel — and the rest of the store — through ounie_list_apps → ounie_describe_app → ounie_invoke_app.

{
  "mcpServers": {
    "ounie": {
      "url": "https://mcp.ounie.com"
    }
  }
}

Scoped endpoint — just this app. Point a client here when the agent should only ever see Seller Intel's tools:

{
  "mcpServers": {
    "seller-intel": {
      "url": "https://seller-intel.ounie.com/api/mcp"
    }
  }
}

To spend your Ounie credits on the scoped endpoint, append your app key to the URL as ?api_key=… — the header-less form clients like the Ounie AI Team need.

Niche sweep	200 credits per call
Demand refresh (per ASIN)	35 credits per call
Review report	1900 credits per call
Guard pass	40 credits per call
Arbitrage map run	9900 credits per call
Listing audit (per ASIN)	25 credits per call

For agents

Credits used