Developers and AI assistants
Ask your AI assistant which products make money
Opteno has a Model Context Protocol (MCP) server at https://mcp.opteno.app/mcp and a REST API at https://api.opteno.app/api/v1. Connect Claude, or any MCP client, and ask in plain language which products to scale or cut; the assistant answers from the same per-product profit figures the dashboard shows. Access is by workspace API key or OAuth, limited to the stores and scopes you grant.
The MCP server
The server speaks the Model Context Protocol over streamable HTTP at https://mcp.opteno.app/mcp (revision 2025-06-18). Any MCP client can connect: Claude on the web and on the desktop, and any tool that supports remote MCP servers.
Every tool reads the same services the dashboard reads, so an answer and the dashboard can never disagree. Every tool checks the store against the credential’s allow-list and the scope it needs before it runs, and a question about a product that does not exist is answered as a plain fact rather than an error.
Step 1: Connect with OAuth (recommended for Claude)
In Claude, add a custom connector with the URL https://mcp.opteno.app/mcp. Claude registers itself, sends you to Opteno to sign in, and you choose the workspace, the stores and the scopes it gets. Withdraw it at any time under Settings → Developer → Connected apps.
Step 2: Or connect with an API key
Create a key under Settings → Developer → Keys, then paste https://mcp.opteno.app/mcp?key=YOUR_KEY as the connector URL — or keep the key out of the URL with the mcp-remote bridge and an Authorization: Bearer header.
Step 3: Ask
For example: “Which products should I scale or kill this week?”, “Show my top 10 products by ROAS over the last 30 days”, or “Tag my best performers this month as winners”.
The twelve tools
find_store_id
Look up a store’s ID by its name — the ID every other tool needs.
get_store_summary
A store’s totals for a period: ad cost, clicks, conversions, revenue, refunds, net revenue, cost of goods, net profit, ROAS, CTR and CPC, plus its top and bottom five products.
search_products
Search and filter products by name, Shopify ID, status or ROAS, sorted by net revenue, ad cost, ROAS or clicks.
get_product_performance
One product’s metrics by name or Shopify product ID, for 7, 14, 30 or 60 days.
get_variant_breakdown
Per-variant performance for one product — size by size, colour by colour.
compare_products
Several products side by side on cost, revenue and profitability.
get_kill_scale_recommendations
Every product with enough spend sorted into scale, watch or kill by its ROAS, against thresholds you can set.
list_tags
The product tags in a store.
get_products_by_tag
All products carrying a tag, with their metrics.
tag_products
Add a tag to products, creating it if needed. Needs the products:write scope.
remove_tag_from_products
Take a tag off products. Needs the products:write scope.
apply_dashboard_filters
Filter the merchant’s own dashboard — for example to active products with ROAS above 3 — while it is open.
The REST API
The public API lives at https://api.opteno.app/api/v1 and is documented, with an OpenAPI schema, at https://api.opteno.app/api/v1/docs/. Authenticate every request with a workspace API key sent as Authorization: Bearer opt_live_….
GET /api/v1/stores/ lists the stores a key can reach. GET /api/v1/dashboard/products/?store_id=… returns the dashboard’s product rows with 7, 14 and 30-day metrics, summary totals and offset/limit pagination, with the dashboard’s filters: status, country, tags, product type, and minimum and maximum ROAS, POAS, net profit, gross margin, price and clicks.
- Keys carry scopes — dashboard:read, products:write, exclusions:write, markets:write — and an optional store allow-list.
- A key is shown once when it is created, can be revoked at any time, and stops working the moment it is revoked.
- Requests are rate-limited per key, and every response says how much of the budget is left.
- Neither the API nor the MCP server can reach billing, account settings or another workspace.
Questions
Does Opteno have an MCP server?
Yes. It is at https://mcp.opteno.app/mcp and offers twelve tools for store summaries, product and variant performance, product comparisons, scale-or-kill recommendations, tags and dashboard filters. Connect with OAuth or a workspace API key.
Can Claude read my Shopify and Google Ads profit data through Opteno?
Yes, once you connect it. Add https://mcp.opteno.app/mcp as a custom connector in Claude, sign in to Opteno, and choose which stores and scopes Claude may use. It reads the same per-product profit figures the dashboard shows.
Can the AI assistant change my ads or my store?
Only what you grant, and only two kinds of write exist over MCP: adding or removing product tags inside Opteno, and filtering your own dashboard view. It cannot change bids, budgets, campaigns or prices.
Is there a REST API?
Yes, at https://api.opteno.app/api/v1, with interactive documentation at https://api.opteno.app/api/v1/docs/. It returns stores and per-product performance, authenticated by a workspace API key.
See which products are losing you money
Connect Shopify and Google Ads in under five minutes. From $14.99 a month, cancel whenever you want, and no change to your store or your campaigns without your approval.