Feed Audit (Product Feed Hygiene)

io.github.theharrisontruong-web由 GitHub 帳號 theharrisontruong-web 發布v1.0.0更新於 Nov 13, 2026

Audits a product feed (XML/CSV/JSON) for Google Shopping disapproval errors. $1.00/call via x402.

發布方自己的服務：audit.domains - Domain Valuation

已驗證Streamable HTTP可網頁執行資料與分析商業與電商

服務詳細資訊

識別碼: io.github.theharrisontruong-web/feed-audit
傳輸: Streamable HTTP
執行環境: Web executable
信任: Verified
發布於: Nov 13, 2026

安全提示

MCP 服務會接收對話中的工具呼叫請求，並可能執行外部操作。啟用前請先了解該服務能做什麼，並只授予它所需的權限。

註冊中心已驗證此條目所屬的 GitHub 帳號 theharrisontruong-web。這並不表明該帳號就是產品方。

SourceWeft 的檢查只涵蓋條目是否完整、端點是否回應；不審查伺服器的程式碼，也不審查其工具的行為。

在 README 中新增徽章

向讀者展示該 MCP 服務已收錄於 SourceWeft，徽章連結到本頁。

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/io.github.theharrisontruong-web%2Ffeed-audit.svg)](https://sourceweft.com/mcp/io.github.theharrisontruong-web%2Ffeed-audit?utm_source=readme&utm_medium=badge)

概覽

AI 產生的概覽

遠端 MCP 伺服器，掃描 Google Shopping 風格的商品 Feed（XML、CSV 或 JSON），回傳健康度評分、等級與主要缺口。

可以做什麼: 對 Feed 網址執行完整健康度掃描，包含 20 項加權檢查，涵蓋解析有效性、必要屬性、GTIN/MPN 等。
執行免費精簡掃描，僅回傳總體結果：評分、A-F 等級、遭拒風險等級、偵測到的格式、商品數量與主要缺口標題。
從完整掃描取得商品層級明細，查看哪些商品未通過哪些檢查。
適用情境: 適合維護 Google Shopping 或類似通路商品 Feed 的情境，在發布前後快速自動檢查遭拒風險。精簡掃描適合初步了解，完整掃描用於商品層級診斷。
執行需求: 需要遠端 streamable-HTTP 端點；無需本機安裝、帳號或 API 金鑰。完整 audit_feed 呼叫透過 x402 協定按次付費，使用 Base 主網上的 USDC，因此需要支援 x402 的用戶端與已儲值錢包；audit_feed_lite 免費。

安裝前請注意

完整掃描按次收費（Base 主網 USDC），且付款即為唯一身分驗證，用戶端會涉及錢包金鑰。伺服器會在伺服器端抓取你提供的 Feed 網址，因此 Feed 內容會傳送給第三方。啟用付費呼叫前請確認價格與付款細節。

安裝

在 SourceWeft 中

開啟儀表板中的 Feed Audit (Product Feed Hygiene)，將其新增到工作區。
為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.theharrisontruong-web%2Ffeed-audit/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "feed-audit-product-feed-hygiene": {
      "type": "http",
      "url": "https://feed-al6bxcngia-ue.a.run.app/mcp"
    }
  }
}

README

Paid Agent APIs — Storefront Grader & GEO Audit

Two pay-per-call APIs built for AI agents. No accounts, no API keys, no subscriptions — payment is the auth. Each call is paywalled with the x402 protocol: send a request, get an HTTP 402 with payment requirements, pay in USDC on Base, get the result.

Operated by Harrison Truong · Contact: [email protected]

1. AI Storefront Readiness Grader — $1.00 / call

Grades any storefront URL A–F across 20 weighted conversion-readiness checks (checkout friction, trust signals, mobile readiness, page performance basics, merchandising completeness, and more) and returns the grade plus the failing checks.


HTTP endpoint	`POST https://grader-al6bxcngia-ue.a.run.app/grade`
MCP endpoint	`POST https://grader-al6bxcngia-ue.a.run.app/mcp`
MCP tool	`grade_storefront`
Input	`{ "url": "https://your-store.example" }`
Price	$1.00 per call (USDC, Base mainnet)
Directory listing	https://www.x402scan.com/server/47e39cb6-7143-4052-b4b1-4dac0e1a2b40

2. GEO Audit — $1.50 / call

Scores any URL's visibility to AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews) across 20 weighted checks — crawlability for AI bots, structured data, content extractability, entity signals — and returns the score plus concrete fixes.


HTTP endpoint	`POST https://geo-al6bxcngia-ue.a.run.app/geo`
MCP endpoint	`POST https://geo-al6bxcngia-ue.a.run.app/mcp`
MCP tool	`audit_geo`
Input	`{ "url": "https://your-site.example" }`
Price	$1.50 per call (USDC, Base mainnet)
Free tier	`POST https://geo-al6bxcngia-ue.a.run.app/geo-lite` — tool `audit_geo_lite`, free, rate-limited to 20/hour

Payment terms (x402)

Protocol: x402 (HTTP 402 Payment Required), exact scheme
Network: Base mainnet (eip155:8453)
Asset: USDC (0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913)
Pay to: 0xd702C8A8Ee30F6950A2a663010d12FC195D5348e
Facilitator: PayAI (https://facilitator.payai.network)
Discovery: both services serve a discovery document at /.well-known/x402 and declare the Bazaar discovery extension (extensions.bazaar) in their 402 responses.

Any x402-aware client works out of the box (@x402/fetch, x402 Python package, CDP or PayAI client wrappers). Unpaid requests receive a standard 402 challenge with the full payment requirements — no signup needed to inspect them:

bash

curl -i -X POST https://grader-al6bxcngia-ue.a.run.app/grade \  -H 'Content-Type: application/json' \  -d '{"url": "https://example.com"}'# → HTTP 402 with payment requirements (1.00 USDC, Base mainnet)

curl -i -X POST https://grader-al6bxcngia-ue.a.run.app/grade \  -H 'Content-Type: application/json' \  -d '{"url": "https://example.com"}'# → HTTP 402 with payment requirements (1.00 USDC, Base mainnet)

Run a paid call from code (buyer example)

examples/buy-with-x402/ is a runnable Node script that buys a call end-to-end with @x402/fetch. It defaults to a dry run: it sends one unpaid request, prints the decoded 402 challenge, and exits — it only pays when you pass --pay with your own wallet key in EVM_PRIVATE_KEY.

bash

node examples/buy-with-x402/buy.mjs            # dry run, prints the 402node examples/buy-with-x402/buy.mjs geo https://your-site.example

node examples/buy-with-x402/buy.mjs            # dry run, prints the 402node examples/buy-with-x402/buy.mjs geo https://your-site.example

MCP usage

Both services expose a stateless streamable-HTTP MCP endpoint at /mcp on their base URLs. Add either to any MCP client as a remote server:

json

{  "mcpServers": {    "storefront-grader": { "url": "https://grader-al6bxcngia-ue.a.run.app/mcp" },    "geo-audit": { "url": "https://geo-al6bxcngia-ue.a.run.app/mcp" }  }}

{  "mcpServers": {    "storefront-grader": { "url": "https://grader-al6bxcngia-ue.a.run.app/mcp" },    "geo-audit": { "url": "https://geo-al6bxcngia-ue.a.run.app/mcp" }  }}

Tool calls return an in-protocol 402 until the client pays via x402; the free audit_geo_lite tool on the GEO server works without payment.

Registry metadata in this repo

server.json — Official MCP Registry manifest for the Storefront Grader (remote).
server-geo.json — Official MCP Registry manifest for GEO Audit (remote).
glama.json — Glama maintainer claim for this repository.

License

MIT — see LICENSE. The license covers this documentation/metadata repo; the hosted services themselves are pay-per-call as described above.

來源：README.md，提交 cca6802

工具

audit_feed

Full product feed hygiene scan of a product feed URL (Google Shopping-style XML, CSV, or JSON — format auto-detected). Fetches the feed server-side and runs 20 weighted checks: parse validity and encoding, required attributes (id, title, link, price, availability values, image_link), GTIN/MPN and brand coverage, identifier_exists misuse, title/description quality, image-link spot-checks, price format and sale_price logic, taxonomy and shipping/tax presence, duplicate ids and variant grouping. These are the errors behind Google Merchant Center disapprovals — a disapproved feed means the products do not serve on Google Shopping at all, and agent-commerce surfaces referencing Merchant Center IDs inherit the same dirt. Returns machine-readable JSON: 0-100 score, A-F grade, a disapproval-risk level (Low/Moderate/Elevated/High) with the failing disapproval-class checks named, per-check pass/partial/fail/unknown with evidence, the worst-offending items sampled with concrete fixes, and the top 3 prioritized fixes. HONEST LIMITS: the scan validates the feed document (capped at 8 MB, first 20,000 items, 8 image spot-checks); it does not crawl landing pages, so feed-vs-page price/availability mismatches are out of scope. Cost: $1.00 USDC on Base per call, paid via x402 — the caller's agent pays from its owner's approved wallet; no API key needed. Use this after every catalog change, before a Shopping campaign launch, or to triage a disapproval wave.

輸入參數url必填

audit_feed_lite

FREE lite product feed hygiene scan of a product feed URL (XML, CSV, or JSON — auto-detected). Runs the same 20 checks as audit_feed but returns top-line results only: 0-100 score, A-F grade, disapproval-risk level, detected format, item count, check counts, and the titles of the top gaps — item-level detail is capped (no per-check evidence, no offending items, no detailed fixes). Use this to triage whether a feed needs the full audit_feed report, or call audit_feed directly for the complete evidence-backed scan.

輸入參數url必填