Run your WooCommerce store by chatting with Claude: sales reports, low stock, orders, bulk price edits and coupons — with preview before every change and one-click undo.
Scored across 25 tools
Most tools are clearly separated by resource and action, such as find_orders vs get_order and update_products vs apply_product_content. There is some overlap among analytics/reporting tools and product diagnostics, but the descriptions give enough scope detail to choose correctly.
Names consistently use snake_case, with action tools mostly following verb_noun patterns and report tools using noun phrases. The two conventions are semantically predictable, though the mix of find/get/list and noun-only report names is a minor deviation from perfect consistency.
At 25 tools, the set is heavy but still plausible for a broad WooCommerce copilot covering analytics, products, orders, coupons, and store health. Some reporting and diagnostic tools could likely be consolidated, making the count borderline rather than well-scoped.
Core workflows are well covered: product search/update/content, order lookup/update, coupon creation, sales/customer analytics, brand voice, and scheduled changes. Gaps exist around product creation/deletion, coupon lifecycle management, and refunds, but many can be worked around in WooCommerce itself.
Saves new names, descriptions or short descriptions you (the AI) have written for specific products. Call get_brand_voice first and follow the user's saved style. Preview first (shows before/after), then confirm. Always show the preview to the user and get their explicit approval before calling again with confirm=true and the preview_id. Paid plans only. SEO title and meta description are saved in Yoast SEO's format (or Rank Math's if the product already uses it). To apply it later instead of now, call with confirm=true, the preview_id AND apply_at (the store's local time, YYYY-MM-DDTHH:MM).
Inputs: items, store, confirm, apply_at, +1 more.
Cancels a change that hasn't run yet. Nothing in the store was touched, so there is nothing to undo.
Inputs: id (required), store.
Lists recent changes Gridly made to the user's stores, with ids that undo_change accepts.
Inputs: limit, store.
Compares two periods side by side with % change in revenue, orders, average order value and refunds. If period_b is empty, compares with the period right before period_a.
Inputs: store, period_a, period_b.
Creates a coupon (percent or fixed amount) with optional expiry, usage limits and category restriction. Preview first, then confirm. Always show the preview to the user and get their explicit approval before calling again with confirm=true and the preview_id. Paid plans only. To apply it later instead of now, call with confirm=true, the preview_id AND apply_at (the store's local time, YYYY-MM-DDTHH:MM).
Inputs: code, store, amount, confirm, +8 more.
Customer counts, new vs returning customers (looking back a year), repeat-buyer rate and top customers by spend for a period. Names are shortened for privacy.
Inputs: store, period.
Groups customers into VIP (top 10% by spend), repeat, one-time, new (first order in the last 30 days) and lapsed (no order for 90+ days). Names are shortened and emails are never shown. Pass segment to list who is in one group.
Inputs: limit, store, segment, lookback_days.
Finds orders by status, date period, customer email, product, amount, or age (older_than_days). Customer names are shortened (e.g. 'Anna K.') unless you filter by customer_email. Use get_order for full details of one order.
Inputs: limit, store, period, status, +5 more.
Searches products by name, SKU, category, tag, stock, price or status, and can find products with missing data (image, description, price) or thin descriptions (max_description_words).
Inputs: sku, tag, limit, store, +8 more.
Returns the user's saved writing styles (brand voice presets). Call this BEFORE writing product names, descriptions or SEO text, and follow the style. With no name it returns all presets.
Inputs: name.
Full details of one order: items, totals, status, customer, shipping address and notes. Use when the owner asks about a specific order.
Inputs: order_id (required), store.
Revenue, number of orders, average order value, refunds, items sold, payment methods and order statuses for a period. Use for 'how were sales this week/month?'.
Inputs: store, period.
Lists changes waiting to be applied later, plus the most recent ones that ran.
Inputs: no named parameters.
Lists the WooCommerce stores connected to this Gridly account, with currency and status.
Inputs: no named parameters.
Lists published products that are out of stock or have stock at or below a threshold.
Inputs: store, threshold.
Revenue, cost and margin for the products sold in a period. Needs cost prices stored in WooCommerce (its Cost of Goods Sold field or a cost-of-goods plugin). Products without a cost are left out and counted.
Inputs: limit, store, period.
One table with revenue, orders, average order, refunds and top product for each connected store over the same period. Built for agencies; also works with one store. Currencies are never added together.
Inputs: store, period, stores.
Estimates revenue for the next days or weeks by extending the recent weekly trend. A rough guide, not a promise: it ignores seasonality and promotions. Needs at least two weeks of past orders.
Inputs: store, history_days, horizon_days.
Saves (or replaces) a named writing style, for example 'Friendly' or 'Luxury', as plain instructions: tone, words to use or avoid, typical length. Saved in the user's Gridly account only — nothing in the store changes. Set remove=true to delete a preset.
Inputs: name (required), remove, guidelines.
A checklist of what to fix: products missing images, descriptions or prices, out-of-stock and low-stock items, drafts, and orders stuck unpaid, on hold or unprocessed for over 3 days.
Inputs: store.
Looks at recent sales and stock and proposes discount campaigns (clear slow stock, grow the basket, restock best sellers, win back lapsed customers). Each idea includes ready-to-use settings for create_coupon or update_products — show them to the user, then use the normal preview and confirm steps.
Inputs: store.
Ranks products by revenue or units sold in a period. 'bottom' lists the slowest sellers among products that sold at least once.
Inputs: by, limit, order, store, +1 more.
Reverses a change Gridly made earlier (see change_history for ids). Preview first, then confirm. Fields that were edited again after the change are left alone and listed. Always show the preview to the user and get their explicit approval before calling again with confirm=true and the preview_id.
Inputs: store, confirm, change_id, preview_id.
Changes an order's status and/or adds a note (private, or visible to the customer). Preview first, then confirm. Always show the preview to the user and get their explicit approval before calling again with confirm=true and the preview_id. Refunds must be done in WooCommerce itself. To apply it later instead of now, call with confirm=true, the preview_id AND apply_at (the store's local time, YYYY-MM-DDTHH:MM).
Inputs: note, store, status, confirm, +4 more.
Changes prices, sale prices, stock, status, categories or tags for many products at once. Works in two steps: a preview shows exactly what would change (nothing is changed), then confirm=true applies it and records it so it can be undone. Always show the preview to the user and get their explicit approval before calling again with confirm=true and the preview_id. Max 500 products per change. Paid plans only. To apply it later instead of now, call with confirm=true, the preview_id AND apply_at (the store's local time, YYYY-MM-DDTHH:MM).
Inputs: ids, sku, tag, price, +13 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Add one secure layer between your agents and this server.