MCP App Store
by KeenEthics
Gaia Suite icon

Gaia Suite MCP Server for ChatGPT

By GaiaSuite

Overview

Gaia helps ecommerce operators inspect their company’s catalog, channels, orders, invoices, fulfillment, returns, and revenue metrics through read-only data lookups.

How to connect

  1. ChatGPT: Open the Gaia Suite listing in the ChatGPT app directory and choose Connect. ChatGPT listing Gaia Suite exposes 33 tools here.

Tools

get_annotations

ChatGPT
One row per comment/event attached to an order, return, or return line, newest first; requires either id or both object_type and object_id — never call with none of the three

get_brands

ChatGPT
One row per brand, alphabetical; use to discover valid brand_id values before calling get_products

get_categories

ChatGPT
One row per category, alphabetical by full path; full_name is the hierarchical path, name the localized leaf label — filter each independently

get_channels

ChatGPT
One row per channel (storefront/marketplace), alphabetical; the returned id is the channel_id every other tool's channel_id filter takes (get_products, get_orders, get_revenues, get_returns, get_proformas, get_product_activations, and more)

get_company

ChatGPT
Return basic information about the current company.

get_countries

ChatGPT
One row per country — global ISO reference data, not scoped to any company; use this to resolve a country name or ISO code, e.g. before setting an address country_code on get_orders, or to look up get_vat_rates' country_code

get_currencies

ChatGPT
One row per currency — global ISO-4217 reference data, not scoped to any company; use this to resolve a currency name or code, e.g. before calling get_exchange_rates, or to interpret get_company's default_currency

get_customers

ChatGPT
One row per customer, newest first; use to resolve the customer_id returned by get_orders(view="full")

get_exchange_rates

ChatGPT
One row per day for one currency pair — global market data, not scoped to any company; newest day first; rate is units of currency per 1 unit of to_currency (amount_in_to_currency = amount_in_currency / rate); to get the rate in effect on a given day, pass day_max as the day after it with limit=1 and the default asc="DESC" — the same lookup the backoffice itself uses

get_languages

ChatGPT
One row per language — global ISO reference data, not scoped to any company; use this to resolve a language name or code, e.g. to interpret get_company's default_language

get_locations

ChatGPT
One row per location; use to resolve the location ids get_products' stock_locations array references. Restricted to production locations only

get_orders

ChatGPT
One row per order (id always present), newest-placed first by default; use get_customers to resolve a customer_id, get_annotations for order comments/events, and get_returns/get_returns_report for return detail

get_pricing_rules

ChatGPT
One row per pricing rule: a discount-by-criteria ruleset a channel can use when its pricing_mode is "pricing_rule" (see get_channels); includes channels[] (the channels currently using it) and lines[] sorted by rank, each with its own discount/rank and required_brand_ids/required_category_ids/required_tag_ids/required_product_ids/required_attributes match criteria (use get_product_attributes to resolve a required_attributes entry's product_attribute_id)

get_product_activation_history

ChatGPT
One row per archived (past) activation status change for a (product, channel) pair, newest first by default; reasons holds any activation/deactivation note (null when there is none) — see get_product_activations for the current state

get_product_activations

ChatGPT
One row per (product, channel) activation attempt, most-recently-changed first by default; "created"=incomplete, "submitted"/"integrated"=under review, "validated"=online, rejected rows carry errors[], and reasons holds any activation/deactivation note (null when there is none)

get_product_attribute_values

ChatGPT
List the distinct values found for one product attribute across non-archived products, with each value's product count (e.g. "SS26": 120, "FW25": 340 for the season attribute).

get_product_attributes

ChatGPT
One row per product attribute defined for the company (e.g. season, color, size, material), ranked order by default; use to discover valid product_attribute_id values before calling get_product_attribute_values. completion says how the value gets filled — a mode plus only that mode's own settings: "imported" = nothing, "manual" = default_value, "smart" = template (or templates, for a composite attribute), "mirror" = mirrored (the source slug it copies), "mapping" = mapped_attribute + template (a source-value -> destination-value map); the three automated modes also carry overridable (whether a user may override the computed value)

get_product_price_history

ChatGPT
One row per archived price change for a product on the given channel, newest first by default; see get_products for the current price

get_product_promotional_price_history

ChatGPT
One row per archived promotional/special price for a product on the given channel, newest by date_start first by default; called "promotional_price" here, "special_price" in the underlying database — see get_products for currently scheduled/active promotional prices

get_product_revenues

ChatGPT
One row per product with order-derived sales/margin metrics over the window, highest revenue first by default; revenue is ex-VAT and prorates the order's shipping/taxes by quantity share (so it sums exactly to an order's own revenue) but net_revenue does not — use get_products for stock filtering instead, not this tool

get_product_rrp_history

ChatGPT
One row per archived RRP (recommended retail price) change for a product in the given currency, newest first by default; RRP is channel-agnostic, so there's no channel_id filter here — see get_products for the current RRP

get_products

ChatGPT
One row per product in the catalog, most-recently-updated first; filter or sort by stock with quantity_available_min/max and order_by="total_stock_available", and use get_product_revenues for order-derived metrics or get_stocks for per-location rows — never a sales aggregate here

get_proformas

ChatGPT
One row per compta (invoice or credit note), oldest first by default; orders lists every linked order as {id, name}, revenue is the ex-VAT amount in the company's own currency, and raw_totals {currency, shipping, taxes, vat, product} holds whichever side type names — invoice or refund, never both — as stored in the linked order's currency; refund_type/refund_comment appear on refunds only

get_returns

ChatGPT
One row per return whose goods have physically come back, newest received first (the only sort); date is the reception date, has_annotations says whether get_annotations has anything for it, and each line carries the litigation thread — see get_rmas for returns still authorized-but-not-arrived, or get_returns_report for aggregate rates

get_returns_report

ChatGPT
Aggregate return metrics for the operations dashboard.

get_revenues

ChatGPT
Get aggregated revenues per date or distribution channel - validated orders only. Returns order count, revenue, net revenue, net margin

get_rmas

ChatGPT
One row per RMA — a return the customer has been authorized to send back but which hasn't arrived — newest first by default; date is when the RMA was raised, and once the goods arrive the row moves to get_returns, which is also where annotations and litigation threads live

get_stocks

ChatGPT
One row per (product, location, delivery batch) stock record, most-recently-updated first by default; unit_cost and value are in the top-level currency, filter by product through product_id (the row still carries the sku), and prefer get_products' full view for a trimmed per-product stock_locations array — use this tool to paginate stock rows directly, to value inventory, or for the quantity breakdown that array omits (ordered/received/defective/returned)

get_stocks_movements

ChatGPT
One row per quantity change applied to a stock row (receptions, sales, adjustments, transfers, returns — reason is free text), newest first by default; diff is the signed delta, quantity is the resulting snapshot right after the movement, not the delta

get_tags

ChatGPT
One row per tag, alphabetical by name; use to discover valid tag_id values before calling get_products

get_vat_rates

ChatGPT
One row per VAT rate — global reference data, not scoped to any company; related to get_countries only by the country_code string (no join between the two tables) — use get_countries to resolve a country name to its code first

show

ChatGPT
Show a Gaia tool's result to the user as an interactive card (list, table, chart). Other tools' results are never shown to the user: call them directly for lookups, ids, data you compute from, and retries. When a result answers the request as it is, call show instead — it runs the tool and returns the same data to you.

show_data

ChatGPT
Show figures you computed yourself to the user: a table, or headline tiles (kind="kpis", one [label, value] row per tile). Use it once, for the final answer; to show a Gaia tool's result, use show.

Capabilities

Interactive

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/gaia-suite"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/gaia-suite)