Suggest API

SuggestAPI - Search Discovery Layer

Turn shopping intent into products, wherever people start.

[Suggest API @suggestapi]

A catalog shouldn’t need separate integrations for shoppers, apps, and AI agents.

Same catalog. Same discovery job.

SuggestAPI separates the surfaces: direct Search Acceleration API for apps, Agent Gateway for MCP + WebMCP.

Full details: https://t.co/DUeE4fsxsu

#mcp #webmcp

[Suggest API @suggestapi]

A catalog shouldn't need separate integrations for shoppers, apps, and AI agents.

Same catalog. Same discovery job.

SuggestAPI separates the surfaces: direct Search Acceleration API for apps, Agent Gateway for MCP + WebMCP.

[Suggest API @suggestapi]

You don't need to rip out Algolia, Typesense, Elasticsearch, Meilisearch, or Shopify search to fix discovery.

SuggestAPI sits in front of what you already run. Same stack, same checkout — the query just gets understood properly first.

[Suggest API @suggestapi]

A citation puts you in the answer. It doesn't close the sale.

After the AI names you, an agent still has to search your catalog, filter by size and budget, and confirm today's price and stock.

Which of those steps breaks most often on your store?

[Suggest API @suggestapi]

Before you worry about AI citations, type this into your own site search:

"waterproof hiking boots for Iceland in October under $250"

Does it keep all 4 constraints, or just "boots"? That's the kind of request agents will send.

[Suggest API @suggestapi]

@samfiguresout Thanks Sam, agreed. The landing side is where we spend our time: taking a request like "hiking boots for Iceland in October under $250" and turning it into constraints the store's existing search can actually use.

[Suggest API @suggestapi]

@noahiglerSEO Seeing the same. Which makes the clicks you do get worth more: AI-referred visitors tend to arrive with a full brief, not a keyword. If site search only matches words, it drops most of that brief on landing.

[Suggest API @suggestapi]

@meteatamel Really useful overview. One gap I'd love to see covered next: discovery before checkout. UCP standardizes the session, but the agent still needs to turn "waterproof boots under $250" into filters the merchant's search understands. That's the layer we work on at SuggestAPI.

[Suggest API @suggestapi]

@sidtechsid Good read. The shift I keep seeing: SEO gets you found, but an agent still has to search your catalog, filter by constraints, compare, and hand off to checkout. Brands that expose that as structured tools (MCP/WebMCP) will be easier to buy from, not just to cite.

[Suggest API @suggestapi]

@marktraphagen And see it fresh. Google's index is a delayed copy of the catalog: a size can sell out at noon and still look in stock in a result. Agents that can query the store's own search get the price and stock right now.

[Suggest API @suggestapi]

@futurebymelissa The hard part is usually keeping the request intact. "Minimal wool coat, warm enough for Toronto, under $400" has 4 constraints, and most store search keeps one keyword. We built SuggestAPI to extract those and query the store's existing search. Happy to compare notes.

[Suggest API @suggestapi]

@dbalcare Great experiment. Feels like budget limits have to be enforced on the agent's side, never trusted from the merchant's tool text. Did any model do better when the price came back as a structured field instead of prose in the description?

[Suggest API @suggestapi]

@moonbrightyan Yes. A citation puts you in the answer, it doesn't close the sale. Worth adding one more check: when that AI visitor lands and types their request into your site search, does it understand them? That's often where the assisted conversion dies.

[Suggest API @suggestapi]

@mattaskumar Agree, and there's a third question after those two: when someone does act on the answer, can they actually buy? A citation names you. It doesn't check today's price or stock. That last step is where a lot of AI demand quietly leaks.

[Suggest API @suggestapi]

Getting cited by AI is step one.

Step two is harder: an AI shopping agent still has to search your catalog, filter it, compare options, and hand off to checkout.

That's the gap SuggestAPI closes. Same catalog. Discoverable and shoppable by agents, not just mentioned by them.

[Marcel van Oost @oost_marcel]

🚨 𝗝𝗨𝗦𝗧 𝗜𝗡: @Mastercard has signed a broad strategic partnership with @OpenAI.

Agentic commerce just got another major infrastructure push.

The partnership includes Mastercard’s Agent Pay, but CEO @MiebachMichael says it goes much further, covering payments, services growth, and AI transformation.

Mastercard is building around four requirements for agentic commerce:

→ Trust in the payment
→ Trust in the AI agent
→ Trust in the ecosystem
→ Merchant reach

And it now has products targeting each layer:

👉 Agent Pay — tokenized payments for AI agents
👉 Verifiable Intent — captures what the buyer actually authorized
👉 Agentic Trust Intelligence — helps banks and merchants assess AI-initiated transactions
👉 Agent Connect — helps merchants become discoverable to AI agents

Shopify is already among the first major deployments, with Shop Pay using Agent Pay and network tokens.

And Mastercard is making one important bet:

𝗧𝗼𝗸𝗲𝗻𝘀, 𝗻𝗼𝘁 𝗼𝗻𝗲-𝘁𝗶𝗺𝗲 𝘃𝗶𝗿𝘁𝘂𝗮𝗹 𝗰𝗮𝗿𝗱 𝗻𝘂𝗺𝗯𝗲𝗿𝘀.

Miebach says agentic commerce is still in its “early days,” but added that agentic transactions are already “ramping up at impressive weekly growth rates.”

Holiday shopping could be the first big test.

The battle to become the payment infrastructure for AI agents is getting VERY real.

Visa, Mastercard, Stripe, PayPal and others are all moving fast.

Who do you think is best positioned to win the agentic payments race?

[Suggest API @suggestapi]

@EverythingPR Makes sense: those retailers expose clean, current product data. Smaller beauty brands can still compete if an agent can query their catalog live and get the right shade, in stock, at today's price, not a delayed copy in an index.

[Suggest API @suggestapi]

@_kekule15 From the ecommerce side: when queries stop being keywords. "316SS 1/2 in valve, 3000 PSI" needs facets, synonyms and ranking control. Even then the engine doesn't read intent, so parsing constraints before the query matters as much as the engine.

[Suggest API @suggestapi]

@redlerz Yes. And when you look at those AI-referred sessions, check what they type into site search after landing. They tend to arrive with long, specific requests. If search returns nothing useful, the citation did its job and the store didn't.

[Suggest API @suggestapi]

@benchiche When the start is an agent, it doesn't browse. It asks the store's search a long, specific question and needs current price and stock back. Most site search was built for 2-word queries. That's the gap to close before 2030.

[Suggest API @suggestapi]

@lancefang7 Big step. Next question is what the search tool returns when an agent asks for a "waterproof trail runner, wide fit, under $150." If store search only matches keywords, structured calls still get weak results. That's the layer we work on at https://t.co/d0NW67N2uw