What is MCP?
The Model Context Protocol (MCP) is an open standard created by Anthropic that allows AI assistants to securely connect to external data sources and tools. The Shogun MCP Server implements this protocol to give AI assistants controlled access to:
Shogun test details and analytics
All Shogun collected data
Test workflow actions like creating, starting, pausing, and stopping tests
Your Shopify store products, collections, pages, templates, and themes
Requirements
An active Shogun A/B Testing account
A Shogun A/B Testing plan with access to the MCP (Advanced plan, Unlimited plan, or Custom Contracts)
An MCP-compatible AI client (e.g., Claude Desktop, ChatGPT, Cursor, etc.)
Installation
The Shogun MCP Server is hosted at: https://ai.shgapi.com
No installation is required - the server is hosted by Shogun and ready to use.
Connecting Claude
Free Claude plans allow one custom connector. Additional custom connectors require a Claude Pro, Max, Team, or Enterprise plan.
Connecting Claude Desktop
Open Claude Desktop and go to Settings > Connectors.
Click the "Add" button, then "Add custom connector"
Set the name to "Shogun" and the URL to
https://ai.shgapi.com/mcp. Leave OAuth settings empty. Then click "Add".
Then, click "Connect" to continue with authentication.
On the next screen, enter your email associated with your Shogun account, click "Sign in" on the confirmation email, then "Authorize" on the final screen. You will then be redirected to Claude Desktop and connected.
When starting a chat, ensure the Shogun connector is enabled.
Connecting Claude Code
Add the Shogun MCP Server using the CLI:
claude mcp add --transport http shogun https://ai.shgapi.com/mcp
The first time you use the server, Claude Code opens a browser window to complete sign-in. Tokens are stored securely and refresh automatically.
Or add it to your .mcp.json file (project root or ~/.claude.json for user-wide access):
{
"mcpServers": {
"shogun": {
"type": "http",
"url": "https://ai.shgapi.com/mcp"
}
}
}
Connecting ChatGPT
ChatGPT doesn't always interpret the data we return as reliably as Claude. For the most accurate results, we recommend connecting Claude instead.
Custom connectors require a ChatGPT Plus, Pro, Team, or Enterprise plan with Developer mode enabled.
In ChatGPT, click your profile name and go to Settings > Apps > Advanced settings.
Turn on Developer mode.
Click "Create app."
Set the name "Shogun" and the URL to
https://ai.shgapi.com/mcp. Leave OAuth settings empty. Check "I understand and want to continue," then click Create.On the next screen, enter your email associated with your Shogun account, click "Sign in" on the confirmation email, then "Authorize" on the final screen. You will then be redirected to Claude Desktop and connected.
Using it in a chat
Start a new chat and click the "+" icon.
Select More > Developer mode.
Choose "Add sources" and enable Shogun.
Connecting Gemini
Custom connectors require a Google Gemini Enterprise plan.
Add the Shogun MCP server using the Gemini CLI:
gemini mcp add --transport http shogun https://ai.shgapi.com/mcp
The first time you use Shogun, Gemini opens a browser window to complete sign-in.
Or add it to your ~/.gemini/settings.json (user-wide) or .gemini/settings.json (project-level):
{
"mcpServers": {
"shogun": {
"httpUrl": "https://ai.shgapi.com/mcp"
}
}
}
Manage the connection from the CLI:
gemini mcp list # View all configured servers
gemini mcp remove shogun # Remove the server
Connecting Other Clients
Any MCP-compatible client (Cursor, Windsurf, Copilot, and others) can connect to Shogun over HTTP. Add a custom or remote MCP server using this URL:
https://ai.shgapi.com/mcp
Most clients support automatic OAuth discovery. Provide the URL above and complete the sign-in flow when prompted, using the email associated with your Shogun account. Check your client's documentation for its specific steps to add a remote MCP server.
Multi-Store Support
When authenticating, you will get access to all Shopify Stores that are associated with your Shogun email.
Ensure each store has access to MCP functionality in order to use it on that store. You can switch between stores via the MCP and work across many stores in a single chat.
If you would like to always use a specific store, add a System Prompt or Custom Instruction to your LLM with these details.
Examples & Best Practices
Use Cases
The Shogun MCP has full access to your A/B Testing environment and is designed to accelerate your test roadmap through improved test analysis and enable automated testing workflows. A full list of capabilities is defined below in the Tools section. Here are some examples of what you can ask the MCP:
"How are my running A/B tests performing?"
"For any running tests that have reached 90% confidence in a winner, promote that winner and end the test."
"Analyze my store analytics and make a few recommendations on what pages to test on next."
"Create a template test. Set the variant to my 'pdp-review-style-variant' template, set the goal to Add to Cart Rate and the traffic split to 50 / 50, and launch the test."
"Create a summary of all tests running across Store A, Store B, and Store C."
"Create a summary of recently ended tests across Store A, Store B, and Store C."
Best Practices
Manually approving Claude for each of the Read-Only Tools can get cumbersome. We are setting the default permission here to "Always allow".
We recommend keeping the "Write/delete tools" set to "Needs approval" to avoid any unwanted action taken on your storefront. Consider granting "Always allow" permission here only once you are highly comfortable and confident working with Shogun's MCP.
Security
Agents will only have access to the Shopify Stores you have access to on your account.
By default, any action taken by agents will require manual approval to proceed.
Shogun provides read and write actions that an agent can take. Any write action an agent takes will be available in the Audit Log of each test for future visibility.
Available Tools
Store Management
list_stores
Lists every store your Shogun account can access, including which one is currently active.
Returns your active store plus a list of stores — name, .myshopify.com domain, primary domain, currency, time zone, and whether A/B Testing is active on each.
switch_store
Changes which store subsequent questions apply to. Only needed if you have multiple stores.
Returns the newly active store's summary. The store must be linked to your account and have A/B Testing active.
Experiment tools
list_experiments
List Shogun AB tests for the active store.
Returns experiment summaries with pagination info.
get_experiment_details
Get normalized experiment configuration.
Returns: full setup — name, status, test type, goal metric, variants with traffic split, notes, timeline, and configuration (sensitive keys are never included).
get_experiment_results
Get experiment metric totals and variant breakdowns.
Returns each tracked metric (label, goal, current value, improvement) overall and per variant.
get _experiment_statistics
Get latest Bayesian statistics for winner determination.
Returns whether the result is statistically significant, confidence level, improvement, the winning/losing variant, win probabilities, and session counts — the numbers behind a ship/kill decision.
get_experiment_orders
Get PII-stripped attributed order aggregates for an experiment.
Returns aggregated order data attributed to the test. Shopper personal information is never included.
Analytics tools
get_business_kpis
Get storewide and optimized-traffic KPIs.
Returns: storewide KPIs, KPIs for traffic in active experiments, and a recent A/B testing performance summary — ideal for "how is my store doing?" and weekly reports.
get_product_analytics
Get product-level analytics from Shogun data.
Returns: per-product revenue, add-to-carts, cart conversion rate, and related metrics.
get_test_recommendations
Rank product testing opportunities deterministically.
Returns products ranked by testing opportunity (based on revenue and cart-conversion gaps), whether each is already being tested, and a suggested test type for each.
Catalog Tools
get_products
List Shopify products with limited fields.
Returns product ID, title, handle, template, and variants with prices.
list_collections
List Shopify collections with limited fields.
Returns collection ID, title, handle, template.
list_pages
List Shopify pages with limited fields.
Returns page ID, title, handle, template.
list_themes
Lists the Shopify themes installed on the active store.
Returns an object with a themes array.
list_templates
Lists the template files in the live (main) theme of the active store, each split into a template name and suffix.
Returns an object identifying the live theme and its templates.
get_shogun_schema
Lets your assistant discover Shogun's conventions — supported test types (page, template, theme, URL redirect, price, product details, checkout, shipping), date-filter rules, and safety guarantees — so it can answer accurately.
Returns a machine-readable capability guide. No parameters.
Management tools
create_experiment
Create a draft experiment. Requires user confirmation.
update_experiment
Update a draft experiment. Requires user confirmation.
start_experiment
Start an experiment. Requires user confirmation.
pause_experiment
Pause an experiment. Requires user confirmation.
resume_experiment
Resume an experiment. Requires user confirmation.
select_winner_and_stop
Select a winner and stop an experiment. Requires user confirmation.
Returns the concluded experiment's details. This is the only way to stop a test — choosing a winner is always an explicit decision. To stop without keeping a variant, select the original/control as the winner.
test_again
Create a new draft from an existing ended experiment. Requires user confirmation.
add_experiment_note
Create or update experiment notes.
Returns the updated notes. The only manage action that doesn't require confirmation, since it never affects a live test.