ext-schema-check

name

ext-schema-check

description

Extract and validate schema.org structured data (JSON-LD) from a web page, with a focus on Google Merchant Center / free-listing product eligibility. Use when the user wants to check a product page's structured data, find missing Product markup fields (name, image, price, brand, sku/gtin, availability), audit JSON-LD, or verify a landing page is "merchant ready" before running Shopping / free listings. Pure Python, no MCP, works offline.

ext-schema-check — schema.org / JSON-LD validation for Merchant Center

Self-contained BDOS extension that fetches a page, extracts every <script type="application/ld+json"> block, and validates Product structured data against the field expectations Google applies to Shopping free listings.

Pure standard library — no MCP server, no pip deps, no browser. Lives under my/, so it survives bdos update.

Use it to answer: "Does this product page have the structured data Merchant Center needs?" Combine it with bdos-merchant-review — the feed tells you what Google ingested; this tells you what the landing page actually exposes.

Language

Talk to the user in their language (PL or EN — match how they wrote to you). Code, logs and returned dicts stay in English.

Extract all JSON-LD blocks from a page

import sys; sys.stdout.reconfigure(encoding='utf-8')
from my.extensions.schema_check import extract
r = extract("https://shop.example.com/product")
if r["ok"]:
    print("blocks:", r["count"], "| types:", r["types"])
    if r["errors"]:
        print("parse errors:", r["errors"])
else:
    print("ERROR:", r["error"])

extract returns ok, url, count, blocks (parsed JSON), types (unique @type values found, incl. inside @graph), and errors (per-block malformed-JSON reports — parsing keeps going).

Validate one product page for Merchant Center readiness

import sys; sys.stdout.reconfigure(encoding='utf-8')
from my.extensions.schema_check import validate_product
v = validate_product("https://shop.example.com/product")
if v["ok"]:
    print("found Product:", v["found"], "| merchant_ready:", v["merchant_ready"])
    print("missing required:", v["missing_required"])
    print("missing recommended:", v["missing_recommended"])
    for issue in v["issues"]:
        print(" -", issue)
else:
    print("ERROR:", v["error"])

validate_product checks required fields (name, image, and an offers object with price, priceCurrency, availability) and recommended fields (description, brand, and a product identifier — any of sku / mpn / gtin*). merchant_ready is True only when nothing required is missing. availability accepts both the schema.org URL (.../InStock) and the short form (InStock).

Validate many product pages at once

import sys; sys.stdout.reconfigure(encoding='utf-8')
from my.extensions.schema_check import validate_many
urls = [
    "https://shop.example.com/p1",
    "https://shop.example.com/p2",
]
for v in validate_many(urls):
    if not v["ok"]:
        print("ERROR:", v["error"]); continue
    tag = "READY" if v["merchant_ready"] else "NOT READY"
    print(f"[{tag}] {v['url']} — missing required: {v['missing_required']}")

Result shapes

• extract → {ok, url, count, blocks, types, errors} or {ok: False, error}.
• validate_product → {ok, url, found, product, missing_required, missing_recommended, issues, merchant_ready} or {ok: False, error}.
• validate_many → list of validate_product results.

Every public function returns a dict with an ok key; on network/parse error it returns {"ok": False, "error": "..."}.

Notes

• A Product node is found even when nested inside a top-level array or an @graph container — the search recurses.
• Fetch uses a browser-like User-Agent, follows redirects, 20s timeout by default (timeout= to change). Some sites still block server-side fetches; a {"ok": False, "error": "fetch failed: ..."} result is expected there.
• Run scripts with the BDOS venv Python (path in the session banner). This extension needs nothing beyond the standard library.
• Pairs well with bdos-merchant-review (feed side) and bdos-feed-optimize (fix titles/descriptions) when a page's markup is thin.