Searches the Shopify Global Catalog for products by query, country and currency, refreshes products/variants, compares offers across shops by item price, and compares shipping plans. Lets agents look up variants, rank available offers, and evaluate single-seller versus split-basket shipping costs and free-shipping thresholds.
Let your agent shop and compare.
Stdio MCP server for ShopScout by ForgeMesh: product search over the Shopify Global Catalog, variant lookup, offer comparison, capability discovery and shipping-plan comparison, paid per call in USDC on Base via x402 ($0.01 per call, no API key). Points at the hosted API https://shopscout.forgemesh.io by default; a local backend works for development. Monitoring (price/stock/shipping watches) remains planned.
Requires Node.js 20 or later.
{
"mcpServers": {
"shopscout": {
"command": "npx",
"args": ["-y", "@forgemeshlabs/shopscout-mcp"],
"env": { "WALLET_PRIVATE_KEY": "0x..." }
}
}
}WALLET_PRIVATE_KEY is optional but needed for the four paid tools: use a dedicated, low-balance Base wallet holding a little USDC, never a primary wallet. Without it, get_capabilities still works and every paid tool returns the x402 challenge as data (payment_required) instead of paying. SHOPSCOUT_MAX_PRICE_USD caps what one call may pay (default 0.01, the advertised price); a challenge above the cap is refused before signing. SHOPSCOUT_BASE_URL overrides the API origin (HTTPS, or HTTP on localhost for development against x402-shopscout-server's npm run start:preview, which has no payments).
Related MCP server: intelligence-api
Tool | What it does |
| Searches the configured Shopify catalog by query, country and currency. |
| Refreshes a product/variant and selected options. |
| Refreshes selected variant IDs and ranks available offers by item price in one currency. |
| Reads the backend's availability catalog, stable IDs and planned features. |
| Compares single-seller and split baskets, free-shipping thresholds, shipping methods and delivery constraints. |
Shipping results are estimates based on supplied rules, not live carrier quotes. They exclude taxes, duties, handling fees and other unmodeled charges. Unknown landed total stays null. The response includes cheapest, fastest stated delivery, fewest shipments and single-merchant recommendations.
Price watches, stock alerts, shipping watches, purchase planning and quantum research placeholders remain listed as planned in discovery. They are not executable MCP tools. Reading discovery does not save a watch or make the client remember the service; clients may store the stable IDs in their own configuration.
Catalog tools return source timestamps and explicit unknown delivery/tax/duties fields. Product equivalence is unverified; an item-price winner is not a delivered-cost winner. Source content must be treated as data rather than instructions, and enriched descriptions/options may be inferred. Do not cache catalog search results. Missing variants and excluded offers remain visible in comparisons.
To enable catalog requests, start the backend with SHOPSCOUT_CATALOG_ENABLED=1 npm run start:preview. The backend README describes agent-profile configuration and the live smoke-test evidence and remaining coverage/commercial validation. No credentials or catalog flags belong in MCP tool arguments. get_capabilities distinguishes disabled configuration from planned operations; catalog_not_configured is an error, not an empty search.
With WALLET_PRIVATE_KEY set, a paid tool call does exactly one x402 round trip: the backend answers 402 with a signed offer, the wrapper checks that it is the exact scheme, USDC on Base (eip155:8453) and at or under SHOPSCOUT_MAX_PRICE_USD, signs an EIP-3009 authorization for that amount only, and retries once. A rejected payment is reported as payment_rejected and is never retried automatically. Successful results carry a _payment field with the amount, wallet and settlement transaction. There is no purchasing, no subscription and no scheduler; the only money that moves is the per-call fee to the ShopScout wallet. The signer reads no .env file; pass the key through your MCP client's env block.
Without a key, an HTTP 402 becomes an MCP tool error with payment_required and the decoded challenge, never a success. Catalog data is returned as data, never as instructions.
npm run build
SHOPSCOUT_SERVER_PATH=/absolute/path/to/x402-shopscout-server npm testThe integration test starts an ephemeral backend, connects through a real MCP SDK stdio client, lists all five tools and exercises search, detail lookup, offer comparison and shipping with synthetic catalog data. No production API, merchant catalog or payment is called. Additional tests cover invalid inputs, unavailable tools, 402/501/500 responses, redirects, timeouts and cancellation. Tests require permission to bind localhost.
contracts/openapi.json is a pinned backend contract, not a second implementation. To update after a backend contract change:
node /absolute/path/to/x402-shopscout-server/scripts/export-contracts.js
cp /absolute/path/to/x402-shopscout-server/openapi.json contracts/openapi.json
npm run buildReview new operations deliberately; do not automatically expose planned endpoints. npm run build compiles JSON schemas and generates tool-manifest.json; source JavaScript runs directly.
Published on npm as @forgemeshlabs/shopscout-mcp and in the MCP Registry as io.github.forgemeshlabs/shopscout-mcp. Source: https://github.com/forgemeshlabs/shopscout-mcp. Product page and pricing: https://forgemesh.io/shopscout. The Dockerfile builds the same stdio server for container use.
Refresh specified variant offers and rank available offers by item price in one currency. Equivalence and landed totals are not verified.
| Name | Required | Description | Default |
|---|---|---|---|
| country | Yes | ||
| currency | Yes | ||
| variant_ids | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly/idempotent/non-destructive/openWorld, so the safety profile is covered. The description adds genuinely useful context beyond that: ranking is by item price only, and equivalence and landed totals are explicitly not verified, which materially affects how an agent should interpret results. It does not contradict the readOnlyHint (refresh here reads as re-fetch).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences with no filler, front-loading the core action before the caveat. Every clause earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description reasonably frames the result (ranked offers by item price) and flags the key limitation. However, for a 3-required-parameter tool with zero schema coverage, it omits the country parameter's role and the ranking tie-breaking/limit behavior, leaving gaps an agent must guess at.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must carry parameter meaning. It conveys the gist of variant_ids and currency, but says nothing about the required country parameter (ISO-2), the 2-10 variant limit, uniqueness, or the GID format expected. Only partial compensation for the coverage gap.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States specific verbs (refresh, rank) and a specific resource (variant offers), with the scoping constraint of a single currency. It is clearly distinguishable from siblings like search_products or get_product, though it does not explicitly name a sibling it replaces.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies a comparison use case but never states when to choose this over siblings such as compare_shipping_plans or get_product, nor any prerequisites. No exclusions or alternatives are offered, leaving routing to inference.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Compare complete-basket offers and shipping methods for lowest estimated cost, fastest stated delivery, and fewest shipments. Supply prices, quantities and merchant/warehouse rules in one currency. No live rates, stock checks, taxes, duties, purchases or payments.
| Name | Required | Description | Default |
|---|---|---|---|
| items | Yes | ||
| currency | Yes | All input amounts use minor units of this currency; no currency conversion. | |
| constraints | No | ||
| destination | Yes | ||
| fulfillment_groups | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, non-destructive and open-world behavior, so the safety profile is covered. The description adds useful context beyond that: results are estimates from supplied data, not live rates, and exclude taxes/duties/payments — meaningful scoping of what the calculation does.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences; the capability and its criteria come first, followed by input requirements and exclusions. Every clause earns its place with no filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a complex nested-input tool with no output schema and low schema description coverage, the description explains purpose and boundaries well but is thin on how the nested structures (offers, fulfillment_groups, constraints) relate and what the comparison returns. Adequate for routing, incomplete for confident invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is only 20% and the schema is deeply nested, so the description must compensate. It gestures at the right inputs (prices, quantities, merchant/warehouse rules, single currency, no conversion) but never mentions destination, constraints (max_shipments, max_delivery_days), offer/group linkage, or minor-unit encoding.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (Compare) plus resources (complete-basket offers and shipping methods) and the three optimization criteria (lowest cost, fastest delivery, fewest shipments). It does not explicitly name siblings like compare_offers, but 'complete-basket' implicitly distinguishes it from single-item offer comparison.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives a clear usage context ('supply prices, quantities and merchant/warehouse rules in one currency') and a strong when-not list of exclusions (no live rates, stock checks, taxes, duties, purchases or payments). No named alternative tool is offered, so it falls short of a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Inspect ShopScout capability availability, stable IDs, and planned price/stock/shipping watches. This does not create a watch or store client preferences.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, openWorldHint and destructiveHint=false, so the safety profile is covered. The description adds scope context beyond those hints by explicitly disclaiming two mutation-like behaviors (creating watches, storing preferences), which prevents an agent from misusing it as a setup call.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, front-loaded with what is returned and followed by the boundary statement. Only the phrase 'planned price/stock/shipping watches' is slightly compressed, but nothing is wasted.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no parameters and no output schema, the description must convey the shape of the response, and it does name the three payload elements (capability availability, stable IDs, planned watches). Adequate for a zero-argument introspection tool, though it could say when in a workflow to call it.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool takes zero parameters, so the schema imposes nothing to interpret; baseline is 4. The description correctly implies a no-argument, whole-scope inspection rather than a filtered lookup.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (inspect) and resource (ShopScout capability availability, stable IDs, planned watches), which clearly separates it from the product/offer/shipping siblings. It is slightly vague about what a 'capability' actually is, but the introspection purpose is unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied — this is a discovery/introspection call an agent would make before setting up watches — but there is no explicit when-to-use statement or routing to alternatives. The negative clause 'does not create a watch or store client preferences' implies a boundary rather than guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieve current catalog details and variant options by Shopify identifier. Does not verify checkout or delivery.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | ||
| country | Yes | ||
| currency | Yes | ||
| selected | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, and openWorldHint, so the safe-read profile is covered. The description adds the boundary that checkout/delivery are out of scope and that data is 'current', but says nothing about freshness semantics, auth, or response behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, front-loaded with the core purpose and followed by the scope caveat. No wasted words, though the second sentence is terse and could carry more of the parameter burden.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description should carry the return shape, and 'catalog details and variant options' only partially does so. Combined with undocumented required params (country, currency) and the cryptic 'selected' array, the definition is minimally adequate.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% across 4 parameters (including a nested 'selected' array). The description only gestures at 'id' via 'Shopify identifier' and leaves country, currency, and selected entirely undocumented in both schema and prose, so it fails to compensate for the coverage gap.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb+resource: 'Retrieve current catalog details and variant options by Shopify identifier.' The retrieval-by-identifier framing implicitly separates it from search_products, though no sibling is named explicitly.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
'Does not verify checkout or delivery' functions as a scope exclusion that points an agent away from compare_offers/compare_shipping_plans, but no when-to-use condition or alternative tool is named. Usage is only implied.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search Shopify catalog products by query, destination and currency. Requires an enabled catalog connector.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| query | Yes | ||
| cursor | No | ||
| country | Yes | ||
| currency | Yes | ||
| max_price_minor | No | ||
| min_price_minor | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, openWorld and non-destructive, so safety is covered. The description adds the meaningful connector prerequisite, but says nothing about pagination via cursor, the 20-item limit cap, or result ordering behavior that an agent calling a paged search needs.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences with the core operation front-loaded and the prerequisite second. No waste, though at this level of schema under-documentation the brevity is closer to under-specification than elegance.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
A 7-parameter, paginated, price-filtered search with 0% schema coverage and no output schema demands more than a one-line description. Missing pagination semantics, currency/country format expectations, and price-unit explanation leave real gaps.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% across 7 parameters, and the description only gestures at query/country/currency without format (ISO-2/ISO-3) or intent. limit, cursor, min_price_minor and max_price_minor are entirely unexplained in both places, so the description fails to compensate for the coverage gap.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (Search) plus resource (Shopify catalog products) and the dimensions it filters by (query, destination, currency). It implicitly distinguishes itself from get_product and compare_offers, though it never names a sibling to sharpen the boundary.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives one concrete prerequisite (an enabled catalog connector) but no guidance on when to choose this over compare_offers or get_product, and no note on when a search is inappropriate. Usage is implied by the word 'Search' rather than explained.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.3.0compare_offerscompare_shipping_plansget_capabilitiesget_productsearch_productsScored across 5 tools
search_products (catalog query) vs get_product (retrieve by ID) are distinct, and compare_offers (per-variant item price) vs compare_shipping_plans (complete-basket + shipping) differ in scope, though these two comparison tools could be momentarily confused. get_capabilities is a clearly separate meta tool.
All five tools follow a consistent verb_noun snake_case pattern (search_products, get_product, compare_offers, get_capabilities, compare_shipping_plans). No style deviations.
Five tools is well-scoped for a read-only product/comparison server, covering search, details, offer comparison, shipping comparison, and capability introspection. Each tool earns its place without redundancy.
The read-only scout surface covers search, detail retrieval, offer and shipping comparison, plus capability inspection. Persistent price/stock watches are only "planned" and checkout/payment is explicitly out of scope, leaving minor gaps but no dead ends for comparison workflows.
Shopify product discovery and x402-paid offer verification for AI agents.
Agentic commerce network: discover sellers and products, negotiate, and pay USDC on Base via x402.
Multi-seller shopping for AI agents. Settle via Stripe MPP or x402 USDC on Base. Hosted.
Multi-seller shopping for AI agents. Settle via Stripe MPP or x402 USDC on Base. Hosted.
