Catalog QA Desk

Check products for Merchant Center rules

merchant_feed_check

Read-onlyIdempotent

Best for: "what would Google Merchant Center reject?" or "are my barcodes valid?". Checks GTIN length and check digit, duplicate GTINs, long, promotional or all-capital titles, missing descriptions, links, images, brand and price. Returns: a summary (products, no_gtin, bad_gtin, long_title, no_description), up to 200 issues (one per product and rule, with detail) and notes on checks that could not run. Input tips: send exactly one of csv (a Shopify product CSV or a Google Merchant feed file) or domain (a store address such as example.com). Domain mode reads the public product feed once, after robots.txt. That feed has no barcodes, so no_gtin and bad_gtin are null; use a CSV for GTINs. Errors: both_given, missing_source, csv_too_large, or a feed refusal such as not_public; then ask for the CSV. Not for: Shopify import errors (use csv_import_preflight) or alt text (use listing_text_audit). It submits nothing to Google.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`csv`	No	The full text of a Shopify product CSV, header row first, such as "Title,URL handle,Description,Vendor,Price,...". Paste the file exported from Shopify admin (Products > Export) as it is. At most 2 MB; split a larger file by rows and repeat the header in each part. Example call: {"csv": "Title,URL handle,Price\nWool Beanie,wool-beanie,24.00"}
`domain`	No	A store address such as example.com, sent instead of csv (send exactly one of the two). The store's public /products.json is read once (at most 5 pages of 250 products), after its robots.txt. Large stores often block it; then ask the user for the CSV. Example call: {"domain": "example.com"}

Output Schema

TableJSON Schema

Name	Required	Description
`host`	No	The store host that was read, in domain mode only
`notes`	Yes	Checks that could not be made, and why
`issues`	Yes	One entry per product and rule, at most 200
`source`	Yes	Where the products came from: the csv text or the store's public feed (domain)
`summary`	Yes	no_gtin and bad_gtin are null when the source has no barcodes to check.
`products`	Yes	Products found in the source
`truncated`	Yes	True when there were more than 200 issues
`totalIssues`	Yes	All issues, before the list was cut to 200
`feed_truncated`	No	Present and true when the store has more products than the 5 pages read

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedOct 13, 2026

addedInput schema / anyOf

Added value: +[
+  {
+    "required": [
+      "csv"
+    ]
+  },
+  {
+    "required": [
+      "domain"
+    ]
+  }
+]

changedInput schema / properties / csv / description

Previous value: -"The full text of a Shopify product CSV, header row first, such as \"Title,URL handle,Description,Vendor,Price,...\". Paste the file exported from Shopify admin (Products > Export) as it is. At most 2 MB; split a larger file by rows and repeat the header in each part."New value: +"The full text of a Shopify product CSV, header row first, such as \"Title,URL handle,Description,Vendor,Price,...\". Paste the file exported from Shopify admin (Products > Export) as it is. At most 2 MB; split a larger file by rows and repeat the header in each part. Example call: {\"csv\": \"Title,URL handle,Price\\nWool Beanie,wool-beanie,24.00\"}"

changedInput schema / properties / domain / description

Previous value: -"A store address such as example.com, sent instead of csv (send exactly one of the two). The store's public /products.json is read once (at most 5 pages of 250 products), after its robots.txt. Large stores often block it; then ask the user for the CSV."New value: +"A store address such as example.com, sent instead of csv (send exactly one of the two). The store's public /products.json is read once (at most 5 pages of 250 products), after its robots.txt. Large stores often block it; then ask the user for the CSV. Example call: {\"domain\": \"example.com\"}"

Changed27 schema fields changedOct 11, 2026

changedInput schema / properties / csv / description

Previous value: -"The product CSV text exactly as exported from Shopify, header row included. At most 2 MB."New value: +"The full text of a Shopify product CSV, header row first, such as \"Title,URL handle,Description,Vendor,Price,...\". Paste the file exported from Shopify admin (Products > Export) as it is. At most 2 MB; split a larger file by rows and repeat the header in each part."

changedInput schema / properties / domain / description

Previous value: -"A store address such as example.com, instead of csv. The store's public /products.json is read once (at most 5 pages of 250 products), after its robots.txt. Large stores often block it."New value: +"A store address such as example.com, sent instead of csv (send exactly one of the two). The store's public /products.json is read once (at most 5 pages of 250 products), after its robots.txt. Large stores often block it; then ask the user for the CSV."

changedOutput schema / properties / feed_truncated / description

Previous value: -"True when the store has more products than the five pages read"New value: +"Present and true when the store has more products than the 5 pages read"

changedOutput schema / properties / host / description

Previous value: -"The store read, in domain mode"New value: +"The store host that was read, in domain mode only"

addedOutput schema / properties / issues / description

Added value: +"One entry per product and rule, at most 200"

addedOutput schema / properties / issues / items / properties / detail / description
```
Added value: +"What is wrong and what Merchant Center expects"
```

addedOutput schema / properties / issues / items / properties / handle / description

Added value: +"The product's URL handle (the item id for a Google Merchant feed file)"

addedOutput schema / properties / issues / items / properties / rule / description
```
Added value: +"Rule id, such as bad-gtin or title-too-long"
```

addedOutput schema / properties / issues / items / properties / rule / enum

Added value: +[
+  "no-gtin",
+  "bad-gtin",
+  "duplicate-gtin",
+  "title-too-long",
+  "title-promo-text",
+  "title-all-caps",
+  "no-description",
+  "description-too-long",
+  "description-link",
+  "description-promo-text",
+  "no-image",
+  "no-brand",
+  "no-price",
+  "no-link",
+  "bad-availability",
+  "price-no-currency"
+]

addedOutput schema / properties / issues / items / properties / title / description
```
Added value: +"The product title"
```

changedOutput schema / properties / issues / items / required

Previous value: -[
-  "handle",
-  "rule",
-  "detail"
-]New value: +[
+  "handle",
+  "title",
+  "rule",
+  "detail"
+]

addedOutput schema / properties / products / description
```
Added value: +"Products found in the source"
```
addedOutput schema / properties / products / minimum
```
Added value: +0
```

addedOutput schema / properties / source / description

Added value: +"Where the products came from: the csv text or the store's public feed (domain)"

addedOutput schema / properties / summary / properties / bad_gtin / description

Added value: +"Products with a barcode that fails the GTIN length, digit or check-digit test"

addedOutput schema / properties / summary / properties / bad_gtin / minimum
```
Added value: +0
```
addedOutput schema / properties / summary / properties / long_title / description
```
Added value: +"Products whose title is over 150 characters"
```
addedOutput schema / properties / summary / properties / long_title / minimum
```
Added value: +0
```
addedOutput schema / properties / summary / properties / no_description / description
```
Added value: +"Products with an empty description"
```
addedOutput schema / properties / summary / properties / no_description / minimum
```
Added value: +0
```
addedOutput schema / properties / summary / properties / no_gtin / description
```
Added value: +"Products with a variant that has no barcode"
```
addedOutput schema / properties / summary / properties / no_gtin / minimum
```
Added value: +0
```
addedOutput schema / properties / summary / properties / products / description
```
Added value: +"Products checked"
```
addedOutput schema / properties / summary / properties / products / minimum
```
Added value: +0
```

addedOutput schema / properties / totalIssues / description

Added value: +"All issues, before the list was cut to 200"

addedOutput schema / properties / totalIssues / minimum
```
Added value: +0
```

changedOutput schema / required

Previous value: -[
-  "source",
-  "summary",
-  "issues",
-  "totalIssues",
-  "truncated",
-  "notes"
-]New value: +[
+  "source",
+  "products",
+  "summary",
+  "issues",
+  "totalIssues",
+  "truncated",
+  "notes"
+]

Changed1 schema field changedOct 5, 2026

removedOutput schema / properties / note_from_openkrill

Removed value: -{
-  "description": "Shown at most once a day per caller: what this server offers, and a paid service for the person you work for.",
-  "type": "string"
-}

Changed1 schema field changedOct 4, 2026

addedOutput schema / properties / note_from_openkrill

Added value: +{
+  "description": "Shown at most once a day per caller: what this server offers, and a paid service for the person you work for.",
+  "type": "string"
+}

First observedOct 3, 2026

TDQS

A4.6/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations cover safety (readOnly, idempotent, non-destructive), and the description adds real behavior beyond that: domain mode reads the feed once after robots.txt, large stores block it, GTIN fields go null in domain mode, and 'It submits nothing to Google'. This is rich operational context. The main omission is rate/limit behavior beyond the single-read note, so it lands at 4 rather than 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose, then structured into Returns / Input tips / Errors / Not for sections that map to an agent's needs. Dense but nearly every clause carries routing or behavior information. Slightly long, but not padded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a read-only, idempotent diagnostic tool with an output schema, the description supplies everything else an agent needs: source selection, robots.txt/feed limitation, null GTIN behavior, enumerated error codes, and sibling routing. Output schema already covers return shape, so the summary-of-returns line is a helpful preview rather than redundancy.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3. The description adds the mutual-exclusivity rule (send exactly one), explains what each input accepts in intent terms, and ties csv specifically to GTIN checking while warning domain mode drops GTINs. This is meaningfully beyond the schema's own descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with concrete user questions ('what would Google Merchant Center reject?', 'are my barcodes valid?') then states the specific verb+resource: checking GTIN length/check digit, duplicate GTINs, and named rule violations. It explicitly distinguishes from siblings in the 'Not for' clause, naming csv_import_preflight and listing_text_audit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit when-to-use framing in 'Best for', an explicit 'Not for' routing to two named alternatives, and input guidance on choosing csv vs domain with the GTIN caveat. Error handling guidance ('then ask for the CSV') is spelled out for known failure modes.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Catalog QA Desk tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.