MCP

SHOPLINE MCP server: connect SHOPLINE to Claude or Codex

TL;DR

SHOPLINE is the platform in this list with the least third-party tooling, which makes an MCP connection disproportionately useful — and it has its own reporting API that answers questions the raw order endpoints cannot.

What already exists for this platform

SHOPLINE has not shipped an MCP server at the time of writing (September 2026), and community coverage is essentially absent.

  • No official server and no meaningful community option — this is the thinnest tooling landscape of any platform here.
  • SHOPLINE has its own reporting API, separate from the order endpoints, which answers sales questions directly rather than by aggregating rows.
  • The API has sharp edges that only appear when you actually call it, several of which are documented below.

What a full MCP server adds

Typical option todayAgent Planners MCP
SHOPLINE toolsNone available13 reads, 11 writes
Native reporting APIn/aUsed, with its definitions stated
Ad spend in the same sessionn/aGoogle Ads, Meta, TikTok on the same key
Approval before a store writen/aQueued for approval, or applied at once on a key minted autonomous; deleting anything, repricing, an absolute stock set, deactivating a code, a discount of 50 % off or more and a product status flip always wait for a person — every write audited
Customer names, email and phonen/aNever from the customer list or the sales report over MCP; the generic Admin read tools return what SHOPLINE sends
Tool counts are derived from the live catalogue rather than typed by hand: this family exposes 13 reads and 11 writes of the 1018 tools on the endpoint. A key only sees the writes if it carries tools:write.

How to configure it

  1. 1Create a key in API & MCP. tools:read for questions; add tools:write if the agent should be able to change things. Tools mode needs a paid plan and fails closed.
  2. 2Connect SHOPLINE under Integrations — through the SHOPLINE app (OAuth), or with a private-app Admin API access token pasted from the store admin. Either way the credential is stored encrypted on our side, which is why the MCP key never carries platform credentials.
  3. 3Add the server to Claude Code, Claude Desktop or Codex using the snippet below.
  4. 4Have the agent call describe_permissions first — it reports the key's mode, scopes, plan entitlement and connected accounts.
  5. 5Call list_accounts for the account ids, and scope each call with one. Resolution is exact or it refuses, naming the candidates.
bash
claude mcp add --transport http agent-plannershttps://www.agentplanners.com/api/mcp \
  --header "Authorization: Bearer ap_live_…"

Claude Desktop and Codex

Claude Desktop takes the same endpoint as JSON:

json
{
  "mcpServers": {
    "agent-planners": {
      "url": "https://www.agentplanners.com/api/mcp",
      "headers": {
        "Authorization": "Bearer ap_live_…"
      }
    }
  }
}
Codex uses the same endpoint from ~/.codex/config.toml, with the key in an environment variable rather than the file: [mcp_servers.agent-planners] / url = "…/api/mcp" / bearer_token_env_var = "AGENTPLANNERS_API_KEY".

What the API does when you actually call it

  • Two sales definitions that must not be mixed. In SHOPLINE's own sales report, net_sales = gross_sales + refunds + discounts (both negative), while total_sales adds shipping and tax. Quoting one and labelling it the other is a real reporting error.
  • Its own attribution capture is sparse — measured at around 11% of orders covered versus 85% when attribution is derived from the order's own ad click ids. Any answer about which channel drove sales has to say which source it used.
  • Conversion rate needs GA4. SHOPLINE has no sessions, so a conversion rate computed from orders alone has no denominator; the honest response is to say so rather than return a number.
  • A fulfilment field can return the string "null" rather than a null value, which passes a naive presence check and then breaks whatever reads it.

Use cases

  • Order-level channel attribution using ad click ids, with the coverage stated so you know how much to trust it.
  • Product profitability after ad spend, joining SHOPLINE orders to the campaigns that drove them.
  • Daily sales anomaly alerts, where a stopped store correctly reads as a zero rather than as missing data.
  • Refund and discount analysis, which is where platform-reported revenue and settled revenue diverge.

Frequently asked questions

Is there a SHOPLINE MCP server?
Not an official one, and effectively no community option at the time of writing (September 2026). A hosted multi-platform server is currently the practical route.
Why can't the agent give me a SHOPLINE conversion rate?
SHOPLINE has no sessions data, so there is no denominator. The correct answer is to say GA4 is needed rather than to divide by something that does not measure sessions.
Which SHOPLINE sales figure should I quote?
Whichever you name explicitly. net_sales is gross sales plus refunds and discounts (both negative); total_sales adds shipping and tax. They are different numbers and mixing them across a report is a reporting error.
Put a human-approved agent on your ad ops

Start free — 2,500 credits a month, no credit card. Reads are free; every write waits for you.

Related reading