[XzAeRosho]

There's a slightly new topic called Agentic Commerce, where you say for example: "purchase for me the most energy efficient dishwasher with a budget of $600", and the agent will connect via specialized via special MCP Servers and APIs to available stores, and will do the full purchase process for you.

This MPP helps bridge the gap between the agent putting the product "in the basket", to actually completing the full purchase process.

Disclaimer: I'm not in any way advocating for this use case, but it's part of my job to understand how it works. Part of what I do is try to help Agents understand, for example, what is "an efficient dishwasher" using actual data, and not hallucinated info.

[seanmcau]

I'm probably overlooking something, but what makes the problem of being able to get from item in basket to item is shipping different from choosing which item(s) to put in the basket?

In other words, if Agents are able to navigate marketplaces, shouldn't that imply they can also navigate a subset of the marketplace, the payment section? Especially given that that section is "easier: theres no need for qualitative (or quantitative) judgement like there is for the shopping portion.

Perhaps its a matter of proper safeguards?

[XzAeRosho]

It's not actually doing browser actions like Playwright or other browser automation tools, rather than direct API and MCP calls/actions. This is a whole new subset of API and connections that are all contained within the Agent context, no browser mocking. That's why they are creating these new protocols, so the full governance can work within the context of the Agent and its available tools.

As I said, it doesn't have to make sense, but this is being pushed on us anyway...

[pythonaut_16]

Thanks for sharing your insights!

It seems like this workflow suffers the same problem as Alexa and Amazon dash buttons: consumers don't typically want the computer to just go buy things for them with no oversight. At least I don't.

Adding a checkout step would make this more plausible to me. "Agent, go find the most efficient dishwasher under $600" where it adds its recommendation to a cart, or even "Find me the best dishwashers under $600" where it creates a catalog page with its recommendations and an easy checkout process with whatever store is actually providing.

[ByThyGrace]

So, what is an efficient dishwasher, in agentic-speak? Furthermore, what is actual data? How is any data you pull remotely a source of trust to answer my question? Surely not just what is on the manufacturer's website?

Not trying to be snarky here, your problem space must be awfully complex.

[whalesalad]

This still does not answer the question. What makes this different from any other API request to Stripe?

[twalla]

As much as I detest having to look at ads or being "influenced" in any way, shape or form, I think the opportunities for exploitation with what you just described is potentially orders of magnitude more harmful. Sure, let me just hand my wallet to a stochastic black box with god-knows-what RL'd biases and then hook it up to adversarial data sources all vying to extract the most money from me - what could possibly go wrong?

[another-dave]

Aside from physical real-world purchases, just opening up the space that agents get access to would be another feature. E.g. if you ask Claude to summarise a Twitter thread, it will say "I can't access, please paste the contents in here". That's fine with a human in the loop, but prevents it using Twitter as a source during deep research, say.

Similarly with paywalled sites like the New Yorker or research journals - If the LLM came back to you and said "I've found these 5 articles. Do you want me to add them as sources to summarise (access cost: $0.05)?" or you give it a budget upfront "Access whatever you think is most useful, but don't spend more than $0.10"

At the moment, sites either allow bots full access or block them, but this could provide a middle ground.