HasData Walmart

com.hasdata由 hasdata.com 发布官方发布者v1.0.2更新于 Oct 24, 2026

Walmart search, product pages and customer reviews on walmart.com and walmart.ca, as JSON.

发布方自己的服务:HasData Airbnb

已验证Streamable HTTP可网页运行网页搜索与抓取商业与电商

概览

AI 生成的概览

HasData 托管的服务器,让助手以 JSON 形式搜索 Walmart、读取商品详情并翻页查看 walmart.com 与 walmart.ca 的顾客评价。

可以做什么
  • 按关键词、分类或完整搜索链接检索商品,支持排序、价格区间、配送方式和筛选条件。
  • 读取单个商品的完整信息:价格、占据购买框的卖家、规格、库存与配送,可选附带其他卖家报价。
  • 按评分、话题、商品状况和已验证购买等条件翻页读取评价,并返回评分分布与 Walmart 的 AI 评价摘要。
  • 以资源形式读取固定取值列表(站点、语言、排序、配送方式),不消耗额度。
适用场景
适合需要公开 Walmart 商品数据来做比价、商品对比、卖家与购买框调研,或分析 walmart.com 与 walmart.ca 评价的场景,且无需 Walmart 开发者或卖家账号。
运行要求
需要一个 HasData API 密钥:远程端点通过 x-api-key 请求头发送,npm 或 PyPI 的 stdio 启动器通过 HASDATA_API_KEY 环境变量提供。需要 MCP 客户端;npm 启动器需 Node.js,Python 启动器需 uv/uvx。需能访问 HasData 端点,调用会消耗额度。
安装前请注意
每次成功调用消耗 10 点额度,开启 otherOffers 再加 5 点;免费额度为每月 1,000 点,付费方案每月 59 美元起。密钥是可计费凭据,不要写入共享配置。工具本身只读,但数据经第三方从公开 Walmart 页面获取;评价含评论者姓名、卖家含商家名称,存储时可能涉及隐私义务。

安装

在 SourceWeft 中

  1. 打开 控制台中的 HasData Walmart,将其添加到工作区。
  2. 连接 x-api-key 凭据。它们会被私密存储,绝不会在公开页面上展示。
  3. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/com.hasdata%2Fwalmart/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "hasdata-walmart": {
      "type": "http",
      "url": "https://mcp.hasdata.com/mcp?apis=walmart",
      "headers": {
        "x-api-key": "<YOUR_API_KEY>"
      }
    }
  }
}

README

Walmart MCP Server

A hosted Walmart MCP server that gives Claude, Cursor, Windsurf and any other MCP client three read-only Walmart tools. Run a keyword or category search, read one item with the seller holding the buy box, and page through its customer reviews, all as structured JSON, with no Walmart developer account and nothing to host.

It reads public Walmart pages that a signed-out visitor can see, on walmart.com and walmart.ca.

1,000 free credits every month, no card required, which is 100 Walmart calls at the 10-credit rate.

https://mcp.hasdata.com/mcp?apis=walmart

[Glama score] [tool contract] [MCP] [Tools]

Contents

What you need

An MCP client and a HasData API key from the dashboard, free to create with no card, and the free tier covers about 100 calls a month at the 10-credit rate. This is a remote server, so the simplest path is a URL and an x-api-key header, with no container to run. A client that only speaks stdio reaches it through a thin launcher, published as @hasdata/walmart-mcp on npm and hasdata-walmart-mcp on PyPI, shown below.

Quick start

The server URL is the same for every client. We run it hands-on in Claude Code and Claude Desktop. The other blocks follow each client's own documented format for a remote server.

FieldValue
URLhttps://mcp.hasdata.com/mcp?apis=walmart
TransportHTTP, streamable
Auth headerx-api-key: HASDATA_API_KEY

Clients with OAuth support can add the same URL as a connector and sign in without putting a key in a config file.

Claude Code
bash
claude mcp add --transport http walmart "https://mcp.hasdata.com/mcp?apis=walmart" \  --header "x-api-key: HASDATA_API_KEY"
Claude Desktop

Settings, then Connectors, then Add custom connector, then paste https://mcp.hasdata.com/mcp?apis=walmart and sign in.

For the config-file route, Claude Desktop loads only local (stdio) servers, so it reaches a remote server through a stdio launcher. The @hasdata/walmart-mcp package is that launcher, and it reads the key from the environment. Add this to claude_desktop_config.json:

json
{  "mcpServers": {    "walmart": {      "command": "npx",      "args": ["-y", "@hasdata/walmart-mcp"],      "env": { "HASDATA_API_KEY": "YOUR_KEY" }    }  }}

For Python instead of Node, swap the launcher for the PyPI package, which uvx runs without a manual install:

json
{  "mcpServers": {    "walmart": {      "command": "uvx",      "args": ["hasdata-walmart-mcp"],      "env": { "HASDATA_API_KEY": "YOUR_KEY" }    }  }}
Cursor

~/.cursor/mcp.json for every project, or .cursor/mcp.json for one:

json
{  "mcpServers": {    "walmart": {      "url": "https://mcp.hasdata.com/mcp?apis=walmart",      "headers": { "x-api-key": "HASDATA_API_KEY" }    }  }}
Windsurf

~/.codeium/windsurf/mcp_config.json. Windsurf calls the field serverUrl, not url:

json
{  "mcpServers": {    "walmart": {      "serverUrl": "https://mcp.hasdata.com/mcp?apis=walmart",      "headers": { "x-api-key": "HASDATA_API_KEY" }    }  }}
VS Code

.vscode/mcp.json in the workspace:

json
{  "servers": {    "walmart": {      "type": "http",      "url": "https://mcp.hasdata.com/mcp?apis=walmart",      "headers": { "x-api-key": "HASDATA_API_KEY" }    }  }}

Example prompts

Each of these lands on one tool, or on two in sequence when the second needs the item id the first returns.

  • Find laptop stands under $30 on Walmart and sort them by price.
  • What does item 18493462688 cost right now, and who holds the buy box?
  • Show me every other seller offering this item and what they charge with shipping.
  • Read the reviews of this item that mention battery life.
  • Pull only the verified-purchase reviews of this item and summarise the complaints.
  • Compare the price of this item on walmart.com and walmart.ca.

A prompt that names a product rather than an item id takes two calls, one search to resolve the id and one product lookup to read it. Reviews work the same way, and the search result carries the id both need.

Tools

ToolWhat it returns
hasdata_walmart_product_getWalmartProductTitle, brand, model, UPC, condition, availability, price with strikethrough before-price and unit price, the seller holding the buy box with its rating and return…. 10 credits a call
hasdata_walmart_reviews_getWalmartReviewsEach review with its rating, title, full text with paragraphs preserved, submission date, author, verified-purchase flag, helpful and not-helpful vote counts, badges,…. 10 credits a call
hasdata_walmart_search_getSearchResultsThe product list with item id, title, product URL, brand, price and strikethrough before-price, star rating, review count, seller, availability, sponsored flag, badges,…. 10 credits a call

Three tools, 10 credits per successful call. Each takes domain, either walmart.com or walmart.ca, and language, where walmart.com serves en and es while walmart.ca serves en and fr. A language the storefront does not offer falls back to its default.

Item ids are storefront-scoped. On walmart.com they are numeric, such as 18493462688, and on walmart.ca alphanumeric, such as 6NZMJ5CW6MH2. An id from one storefront does not resolve on the other.

Get Walmart search results

hasdata_walmart_search_getSearchResults

A page of search results for a keyword, a category, or both.

ParameterTypeRequiredNotes
qstringsee belowThe search term
catIdstringsee belowCategory id from a category URL, such as 976759_1086446_1229651
urlstringA full Walmart search or category URL, scraped as is. Overrides the parameters above
domainstringwalmart.com or walmart.ca
languagestringen, es or fr, subject to the storefront
sortstringbestMatch, priceLowToHigh, priceHighToLow, bestseller, highlyRated or newArrivals
pagenumberResult page, starting at 1
minPrice / maxPricenumberPrice band in the storefront currency
deliveryTypestringshipping or pickup
facetstringOne filter in name:value form, such as brand:Great Value

Send q to search, catId to browse a whole category, or both to search inside one. Neither appears in the schema's required array because either one satisfies the call on its own.

Returns searchInformation, a productResults array, a facets block and pagination. Each result carries position, id, title, url, brand, isSponsored, badges, walmartPlusSavings, categoryPathId, a price object, reviews with rating and totalReviews, image, seller, availability and fulfillment.

The facets block is the map of every filter the query supports, and each value carries the exact string to send back in facet. Running one unfiltered search to read the facets is cheaper than guessing.

json
{  "position": 1,  "id": "18493462688",  "title": "Incipio Portable Foldable Aluminum Laptop Stand and Riser with Adjustable Angles, Anti-Slip and Ventilated Design",  "url": "https://www.walmart.com/ip/Portable-Laptop-Stand-Black/18493462688",  "isSponsored": true,  "badges": ["Overall pick"],  "walmartPlusSavings": true,  "categoryPathId": "4125_4134_1074326_9623037_7875081",  "price": { "currentPrice": 9.96, "currentPriceDisplay": "$9.96" },  "reviews": { "rating": 4.5, "totalReviews": 49 }}

Get Walmart product details

hasdata_walmart_product_getWalmartProduct

One item in full.

ParameterTypeRequiredNotes
itemIdstringsee belowThe Walmart item id
urlstringsee belowA full product URL, scraped as is. Overrides itemId and sets the storefront
domainstringwalmart.com or walmart.ca, ignored when url is given
languagestringen, es or fr, subject to the storefront
otherOffersbooleanAlso collect competing offers. Costs 5 credits on top, 15 instead of 10

Pass itemId or url. As with search, neither is listed as required because either one works alone.

Returns a product object with itemId, title, url, brand, brandUrl, type, model, upc, condition, badges, availability, a price object, the seller holding the buy box, reviews, images, categoryPath, categoryPathId, highlights, specifications, keyItemFeatures, productDetails and fulfillment.

The base call already reports how many competitors the page advertises and the cheapest competing price. Turn on otherOffers only when you need the offers themselves, because it takes a second request to Walmart and costs half again as much.

json
{  "itemId": "18493462688",  "brand": "Incipio",  "condition": "New",  "price": { "currentPrice": 9.96, "currentPriceDisplay": "$9.96", "currency": "USD" },  "seller": {    "name": "Walmart.com",    "id": "F55CDC31AB754BB68FE0B39041159D63",    "returnPolicy": "Free 30-day returns"  },  "reviews": {    "totalReviews": 49,    "rating": 4.5,    "fiveStars": 38,    "fourStars": 4,    "threeStars": 3,    "twoStars": 1,    "oneStar": 3  },  "specifications": [{ "name": "Maximum screen size", "value": "16 in" }],  "fulfillment": {    "type": "FC",    "message": "Pickup, today at Fredericksburg Massaponax Supercenter",    "deliveryDate": "2026-09-09T21:59:00.000Z"  }}

Get Walmart product reviews

hasdata_walmart_reviews_getWalmartReviews

The review feed of one item, ten reviews a page.

ParameterTypeRequiredNotes
itemIdstringsee belowThe Walmart item id
urlstringsee belowA full product URL whose reviews to read. Overrides itemId
domainstringwalmart.com or walmart.ca, ignored when url is given
languagestringLanguage of the review page, not of the reviews themselves
pagenumberReview page, ten a page
sortstringmostRelevant, mostRecent, mostHelpful, highestRated, lowestRated or oldest
ratingnumberKeep one star rating, 1 to 5
aspectIdstringKeep reviews mentioning one topic, by its id
conditionstringKeep reviews about one condition of the item
verifiedPurchasesOnlybooleanKeep only purchases Walmart confirmed

Returns reviewsInformation, a reviewResults array, a filters block, appliedFilters and pagination.

filters is the part worth reading first. It lists the star ratings, frequent mentions and conditions this item can actually be filtered by, each with a count and with the exact value to send back. Given {"name": "Battery Life", "value": "6049", "count": 8} you send aspectId: "6049" and expect eight reviews. Guessing an aspect id instead of reading it here is the usual way to get an empty page.

reviewsInformation carries the item rating, the per-star breakdown, per-aspect scores and Walmart's AI review summary. It also separates totalRatings from totalReviews, which matter separately: the item below has 49 ratings but only 21 written reviews, and paging covers the 21.

json
{  "reviewsInformation": {    "rating": 4.49,    "totalRatings": 49,    "totalReviews": 21,    "recommendedPercentage": 100,    "ratingBreakdown": { "fiveStars": 38, "fourStars": 4, "threeStars": 3, "twoStars": 1, "oneStar": 3 }  },  "reviewResults": [    {      "position": 1,      "id": "434698083",      "rating": 5,      "title": "Good value laptop stand.",      "text": "Good value for money. Not the sturdiest, but that is to be expected for a collapsible laptop stand. Overall gets the job done, I'd buy it again.",      "date": "8/1/2026",      "verifiedPurchase": true,      "helpfulVotes": 0,      "notHelpfulVotes": 0,      "badges": ["Verified Purchase"],      "seller": "Walmart.com",      "language": "English",      "aspects": [{ "id": "284", "polarity": "Positive" }]    }  ],  "filters": [    { "name": "Star rating", "parameter": "rating", "values": [{ "name": "5 stars", "value": "5", "count": 38 }] },    { "name": "Frequent mentions", "parameter": "aspectId", "values": [{ "name": "Sturdiness", "value": "828", "count": 6 }] }  ],  "pagination": { "currentPage": 1, "reviewsPerPage": 10, "totalPages": 3, "totalResults": 21, "nextPage": 2 }}

Prompts and resources

The server exposes 5 resources, one per parameter whose accepted values are a fixed list. Reading one is cheaper than learning the vocabulary from a rejected call, and it costs no credits. Each URI is hasdata://walmart/<parameter>.

ParameterValuesWhat it selects
domain2Walmart storefront the item belongs to. Each storefront has its own catalog, item ids, prices and currency, so an id from one storefront does not resolve on another. Ignored when url is provided. Default is walmart.com.
language3Language of the product details. Availability depends on the storefront - walmart.com serves en and es, walmart.ca serves en and fr. A language the storefront does not support falls back to its default.
sort6Order of the returned reviews, named as the Walmart review page names it. Default is mostRelevant.
sort6The sorting option for the search results.
deliveryType2Keep only the products available with the selected fulfillment method. Default is shipping. Note that pickup is answered against one Walmart store, so an item in stock nationally can still be unavailable for pickup nearby - the response then reports that nothing matched the filters rather than that the query had no results.

The list is served without an API key, so a client can read it before a user has signed up.

Errors and failure paths

Plan for these rather than assuming a happy path.

Two totals disagree in one search response, and both are correct. searchInformation.totalResultsDisplay is the string Walmart prints on the page, such as "1000+", while pagination.totalResults is the number behind it, such as 8005. One is display text and the other is an integer, so do not parse the first or print the second.

Walmart stops serving results after roughly page 10. Beyond that the page comes back empty rather than erroring. A large keyword cannot be enumerated by paging, so narrow it with facet, a price band or a category instead.

Prices belong to one store, and the response says which. searchInformation.storeId names it, and fulfillment.message names it in words, down to "Pickup, today at Fredericksburg Massaponax Supercenter". Comparing prices across calls only means something while that store stays the same.

deliveryType: pickup is answered against a single store. An item in stock nationally can still come back unavailable, because it is unavailable at that one store rather than everywhere.

The product tool reports ratings where you might read reviews. Its reviews.totalReviews is the count of ratings, 49 for the item above, while the reviews tool reports 49 ratings and 21 written reviews separately. Use the reviews tool when the distinction matters.

Review dates are M/D/YYYY strings. "8/1/2026" is the first of August, not the eighth of January. Parse with the format in hand rather than letting a date library guess.

A review can carry an aspect id absent from filters. The block lists the topics the item can be filtered by, which is a shorter list than the topics its reviews were tagged with. Read aspects off the review, and filter only with ids the block offers.

variants is missing rather than empty on an item with no variants. Check for the key before you read it.

Only one condition per request. The parameter takes a single value, so a query across two conditions is two calls.

Results that carry data also carry a requestMetadata.id worth quoting in support.

Pricing, free tier and limits

Each Walmart tool costs 10 credits per successful call. Turning on otherOffers adds 5 credits to the product call, 15 instead of 10, so leave it off unless the competing offers are the point. Response size does not change the price.

The free tier is 1,000 credits every month with no card, which is 100 Walmart calls at the base rate. It renews with the billing cycle, so a low-volume agent runs on the free tier indefinitely.

Paid plans start at $59 a month for 200,000 credits, which is 20,000 calls. The unit price falls with volume, from $2.95 per 1,000 calls on the entry plan to $1.19 on Basic and $0.83 across the Growth tiers. Current figures live on the pricing page.

Your plan also sets concurrency. The free tier allows 5 requests at a time, Startup 5, Basic 15, and the Growth tiers run from 50 to 500. Retry on the 429 with a backoff in anything unattended, because an agent that fans out across item ids will reach the ceiling before you do.

A request that comes back non-200 is not billed. A successful call that finds nothing is still a call.

Tool selection

Start from what the prompt gives you. A keyword or a category goes to the search tool, an item id goes straight to the product or reviews tool. Spending a search call to reach an id you already have is the most common waste.

Then pick by what the question is about. The search result is enough for ranking, price sweeps and share-of-shelf work across many items. The product tool is the only one carrying specifications, the buy-box seller and the competing-offer count. The reviews tool is the only one carrying review text.

Read the filters block before you filter. One unfiltered reviews call tells you which ratings, topics and conditions exist and how many reviews each holds, which turns a guessed filter into a known one.

How it compares

Walmart's own Affiliate and Marketplace APIs are the official routes to this data, and they answer different questions.

Walmart Affiliate APIWalmart Marketplace APIThis server
EligibilityAn approved affiliate accountA Walmart seller accountAn API key
ScopeItems in the affiliate catalogueYour own listings and ordersAny public item page
Competing sellersNot returnedYour own offers onlyThe offer list, with otherOffers
Review textNot returnedReviews of your itemsThe feed, with filters
Buy boxNot returnedFor your itemsWhoever holds it
walmart.caSeparate programmeSeparate accountA parameter
CostFree, when you qualifyFree with a seller accountPaid past the free tier

The row that decides it is scope. Both official APIs answer questions about a catalogue you have a commercial relationship with, which rules them out for watching a competitor. When the items are yours, the Marketplace API is authoritative and free, and you should use it.

FAQ

Is there an official Walmart MCP server?

Walmart does not publish one. This one is maintained by HasData and reads public Walmart pages.

What is a Walmart MCP server?

An MCP server exposes tools an AI client can call. This one turns Walmart search results, item pages and review feeds into JSON an agent can reason over, without a browser or a scraping library in your stack.

Do I need a Walmart account or a seller account?

No. The only credential is your HasData key.

Which storefronts are covered?

walmart.com and walmart.ca. They hold separate catalogues, item ids, prices and currencies, so a cross-storefront comparison is a real comparison rather than a currency conversion.

Why did my item id return nothing?

Most often because it belongs to the other storefront. A numeric id is walmart.com and an alphanumeric one is walmart.ca, and neither resolves on the other. Pass domain to match, or pass the full url and let it set the storefront.

How do I get the competing offers?

Set otherOffers on the product call. Each offer comes back with the seller name, storefront URL, price, condition, shipping cost, delivery date and return policy. It costs 5 credits more, because it takes a second request to Walmart.

Why does the same search return different prices on different days?

Partly because prices move, and partly because the response is answered against one Walmart store, reported as searchInformation.storeId. Hold that store constant before reading a price change as a price change.

Can I use this together with other HasData APIs?

Yes. One key covers everything, and one endpoint serves them all through the apis parameter. Point a client at ?apis=walmart,amazon to get both tool sets in one connection, or at mcp.hasdata.com/mcp for the full catalogue.

Is HasData affiliated with Walmart?

No. HasData is an independent service and is not affiliated with, endorsed by, or sponsored by Walmart. Walmart is a trademark of its respective owner. The tools work with publicly available data only, and you are responsible for using the results in line with Walmart's terms and the law that applies to you.

Compliance and personal data

Reviews carry an author name as the reviewer chose to publish it, along with a verified-purchase flag. Marketplace seller entries carry a business name and a storefront URL. Neither block needs the author fields for sentiment or pricing work, so drop them unless your purpose needs them, and check your own obligations before storing them.

HasData links

Other HasData MCP servers: Google Search, Google Maps, Google Trends, Google Flights, DuckDuckGo, YouTube, TikTok, Instagram, Amazon, Shopify, Yelp, Zillow, Redfin, Airbnb, Booking.com, Indeed.

Development

The launcher is a thin stdio bridge to the remote server, so there is nothing to build.

bash
npm installHASDATA_API_KEY=your_key_here npm test

The tests in test/ assert the tool contract, the part that can break without a commit here. They check that ?apis=walmart returns the expected tool count, that no name changed, that every tool still carries a description, that the either/or parameters this README documents are still in the schema, and that the key in use is actually accepted. That last check calls a tool for real and costs 10 credits, which is the price of a canary that can fail for the right reason.

None of the three tools declares a required parameter, because each accepts one of two inputs. The suite pins the alternatives instead of the required array, which would pass while the schema said nothing.

The contract suite also runs weekly on a schedule, because the upstream tool list can change without anyone touching this repository.

Contributing

A tool table, a response sample or a documented behaviour that does not match reality is worth an issue. There is a template for exactly that. Pull requests are welcome for the same, and for anything in the launcher.

License

MIT, see LICENSE.

来源:README.md,提交 c1c8c68

工具

0
SourceWeft 尚未读取此服务器的工具。

版本历史

3
  1. v1.0.2最新Sep 24, 2026
  2. v1.0.1Sep 22, 2026
  3. v1.0.0Sep 16, 2026

评分

已满足适用检查项的 85%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分

其他 Hasdata 集成