A price without its model, capacity, colour, seller, market, and conditions is difficult to use. This workflow starts with Google Search, then asks you to open the first-party page and record what the page actually shows. The example below is a manual review of Apple’s iPhone 16 purchase page, not a live product feed or an MCP-returned catalog.
A small, dated offer audit that another person can open, inspect, and recheck.
Evidence matrix
Product/modelVariantSeller
Apple iPhone 16128GB BlackApple
Apple iPhone 16128GB WhiteApple
Apple iPhone 16128GB PinkApple
Editorial source audit example
Research workflow: manual source audit2026-09-17
Start with a brief
Turn one input into a usable research task
Find a product page, retain the exact variant and offer language, and leave a dated evidence trail for a human decision.
Stored locally for this task. It is not placed in the setup URL.
Generated prompt
Call spawnr_tools to inspect availability and the exact schema. Then call spawnr_tool_quote with {"tool":"google-search","input":{"q":"site:apple.com/shop/buy-iphone iPhone 16 128GB price","country":"us","language":"en","limit":10}}. Show the quote and wait for approval. Only after approval call spawnr_tool_run with the returned quote_id, approved=true, and the private access_token if returned; never display or export that private token. Treat returned links and snippets as discovery; ask the researcher to open the relevant first-party page manually and record exact model, variant, seller, displayed price, offer condition, currency, capture date, source URL, and fields that still need checking. Leave unknown fields blank. Do not claim a product feed, stock monitor, price history, or automatic comparison.
Choose setup guide
Connect MCP opens the selected setup guide and carries this prompt with it.
Source: open sourceCaptured / reviewed 2026-09-176 records shown
A price row needs its context
The useful unit in ecommerce research is an offer with an identity, not a bare number. In the dated Apple example, the product is Apple iPhone 16, the capacity is 128GB, and the five reviewed colours are Black, White, Pink, Ultramarine, and Teal. The purchase path says “Connect to any carrier later,” and the official page shows `$829.00` for each colour. Those details make the row inspectable: another person can open the same page and see what was recorded.
The page also advertises a carrier deal of up to `$799` after an eligible trade-in. That headline is a different offer path from the unlocked price. Its eligibility and conditions need a separate check. A source audit keeps both paths visible instead of collapsing a conditional credit into the product price.
Exact product and model
Capacity, colour, or other variant attributes
Seller or first-party page
Displayed price and offer wording
Capture date and market
How to run the research
Start with a query that names the product and the attribute that could change the offer, such as capacity, colour, unlocked status, or retailer. Google Search can return up to ten US/English organic results. Treat result titles and snippets as leads. Open the first-party product or pricing page before copying a price, and record the page URL rather than the search result alone.
Then compare the page against the row template. Check whether the model, variant, market, currency, seller, and conditions are visible. If the page presents a discount, trade-in, membership, subscription, or carrier requirement, write that requirement in the offer path. Leave unknown fields blank. A reviewer should be able to tell which fact is visible, which is an interpretation, and which still needs a checkout or policy check.
Use Google for discovery and first-party pages for evidence
Record one row per meaningful variant or offer path
Keep conditional credits separate from base prices
Recheck volatile pages before a purchase or catalog update
What this workflow cannot promise
The Spawn’s Google tool returns a one-time search snapshot. It does not expose a structured ecommerce product feed, stock or availability monitor, historical price series, checkout result, automatic variant extraction, or guaranteed comparison across retailers. The Apple rows are manually reviewed source examples. They are not Google results, not an API catalog, and not a guarantee that the page or price will remain unchanged.
Choose this workflow when the decision benefits from a short evidence trail and a human can open the source. If you need controlled pagination, inventory history, tax and shipping totals, or automated catalog reconciliation, those requirements need a separate product and data contract. Leave the gap explicit instead of filling it with a plausible value.
A captured price can change
Trade-in and carrier offers have eligibility conditions
Search ranking and snippets are not product truth
Unknown stock, tax, shipping, and checkout fields stay unknown
The next useful action
Search for one product and one decision you need to make. Open the official page, fill one row per variant or offer path, and attach the capture date. Before sharing the result, ask a second person to click the source and check the condition that could change the price. That simple review turns a quick search into an auditable ecommerce note.
Search one product
Open the source
Recheck the offer
Questions before you run it
Does this provide a product feed or live price monitor?
Google Search provides a bounded discovery snapshot. The example is a manual first-party source audit with a capture date; it is not a structured feed, inventory check, or price history.
Why keep capacity and colour in the row?
The same model can have different variants and offers. Recording 128GB and the colour alongside the Apple iPhone 16 prevents a price from being detached from the item a buyer can actually select.
Is the up to $799 trade-in credit the product price?
It is a separate Apple carrier deal after an eligible trade-in. The eligibility language and linked terms need review before treating it as a buyer saving.
Can I compare several retailers automatically?
You can search for sources and record rows, but the supported workflow does not guarantee consistent variant matching, tax, shipping, stock, or checkout comparison across retailers.
What does the current Google tool catalog charge?
The current live catalog reports a user price of $0 and requires approval. Availability and pilot allowances are runtime state, so inspect spawnr_tools and the quote before each search.