mcpField
M
MerchantFlow

MerchantFlow

Remote
merchantflow.aiData & Analytics0Updated 2w ago
RepositoryVisit website

About MerchantFlow

MerchantFlow is a read-only profit analytics plugin for Shopify and WooCommerce stores. It combines orders with costs, fees, and ad spend for Claude to query, while an included skill guides tool selection and flags missing or stale data.

Key features

  1. 01Summarize P&L and compare periods
  2. 02Analyze product profit and cost coverage
  3. 03Review ad-channel performance and ROAS
  4. 04Look up cohorts and orders
  5. 05Check integration health and sync status

Use cases

  1. 01Compare this month's P&L with the previous month
  2. 02Identify profitable and loss-making products
  3. 03Assess returns by advertising channel
  4. 04Find products with missing cost data

MCP support

The plugin connects to the MerchantFlow MCP server over HTTP and uses OAuth sign-in. Access is read-only and limited to the authorized MerchantFlow store.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://merchantflow.ai/api/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

MerchantFlow for Claude

MerchantFlow on Smithery

Revenue is easy to see. Profit is not. MerchantFlow joins your Shopify or WooCommerce orders with your product costs (COGS), shipping and 3PL fees, payment fees and Meta, Google and TikTok ad spend. This plugin lets Claude read that data, so you can ask about your store in plain English and get answers built on your real numbers instead of dashboard screenshots.

What you can ask

  • "Summarise last month's P&L and compare it with the month before."
  • "Which five products made the most net profit last month, and which lost money?"
  • "What is my ROAS by channel this quarter, and where should I cut spend?"
  • "How much of my revenue is missing product costs, and which products need them?"
  • "Are any of my integrations out of date or failing to sync?"

Claude picks the right MerchantFlow tool for each question: P&L summaries, product profitability, ad and channel performance, COGS coverage, cohorts, order lookup and integration health. The included merchantflow-financial-copilot skill tells Claude how to route questions, compare periods and flag missing or stale data instead of guessing.

Getting started

  1. Install the MerchantFlow plugin from the Claude plugin directory.
  2. Sign in to MerchantFlow when Claude asks. Don't have an account yet? Start at merchantflow.ai/ai.
  3. Ask a question about your store.

Setup guides for Claude Code, Claude Desktop and other clients are in the MerchantFlow MCP documentation.

Security and privacy

  • The plugin connects to https://merchantflow.ai/api/mcp over HTTP and signs in with MerchantFlow OAuth. No credentials or tokens are stored in this repository.
  • Access is read-only and scoped to the MerchantFlow store you authorise.
  • You can revoke access at any time in MerchantFlow Settings > Developer > MCP.
  • See the privacy policy and terms of service.

Support

Read the troubleshooting guide or email [email protected].

Repository contents

  • .claude-plugin/plugin.json - plugin metadata and directory listing details.
  • .mcp.json - the MerchantFlow MCP connection.
  • skills/merchantflow-financial-copilot/ - tool routing and reporting guidance.
  • assets/logo-icon.png - MerchantFlow brand mark.

This repository contains no MerchantFlow application source code, customer data, credentials or access tokens.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.