A good MCP server makes the difference between an agent using 20k tokens and 2 million. It may not matter yet with sponsored Codex and Claude subscriptions, but it will kill many use cases once providers switch to token-based billing.
That may sound like an exaggeration, but it’s exactly what I see in our product.
Humans developing something already have context that agents don’t have yet. Most agents start a task with virtually no prior knowledge. And they start from zero every single time. That may improve in the future, but we’re not there yet.
Can agents get the job done? Yes. But without a thoughtfully implemented MCP server, they are awkwardly inefficient.
Seems to me that you're saying the MCP is a simplified API with good documentation geared towards agents. But if that's the case, could you not exposes the simplified interface as part of the API, instead of exposing it in MCP?
When it is part of the API, the agent still has the choice between multiple options. If it chooses the less efficient one, the request can become significantly more CPU- and token-intensive than necessary.
The problem is that the agent does not care. Its primary goal is to get the job done.
Maybe the agent is smart enough to choose the optimal path, but that strongly depends on the model being used. You also do not know who is on the other side. With a human-facing API, you can usually assume who is using it and what they want to achieve. Humans are generally lazy and tend to look for the most efficient solution.
An agent, however, will happily iterate through 1,000 users and fetch the online state for each one individually, even across multiple paginated requests if necessary.
You can provide an endpoint that returns the online states for all users at once. A human will most likely use that endpoint, but I have seen agents go completely wild on the other side. :D
At some point, you may get a response like “token limit reached.” But what do you do then? You give the agent more tokens and increase your bill, because you cannot even tell whether there was a more optimized way to achieve the same result.
In practice, this is a surprisingly tricky problem. :D
There are a lot of studies that proof that darkmode during day is straining eyes much more. In the night dark mode is better. But most of the "day-to-day" user are working during day.
Well, it is reversed in a sense, in that it is "no, de" reversed, it's just not reversed character-by-character. I'd agree that's not generally how we used "reversed" for words, but it tracks more generally, you can reverse something without necessarily breaking it down into its smallest possible parts.
For example, that "Pets a dog" reversed could be either "Dog a pets" or "God a step", obviously words are a more natural breakpoint though.
To really try too hard, in Japanese it'd be something like ノデ so it would be reversed character by character as it'd be split like that (not a Japanese speaker, so take with a lot of salt, I imagine it'd actually use another character to sound more like the English pronunciation of node).
Mi name is Florian - UX Designer, Web developer and Hardwareguy - and I stumbled upon a problem at work and tried to find a solution for that:
At work we are using a shared coffeemachine with a coffee tally sheet to track who is drinking coffee. One of my colleagues is buying the coffee and we are paying him each month. He is counting the coffee's per person, sending them an email and often telling them paying strange amounts of money (5.75€ or so).
We thought we can do that better, so I build a simple device that tracks the consumption through our Company ID card. Every user gets notified when the consumption is reaching a predefined threshold (like 5€) to pay the person that's buying the coffee (capsules in our case). It is cloud connected for easy access and can be extended by additional terminals too.
Our department was quite happy with it and other departments requested this solution after they saw our small gadget too. :)
In the future I plan to bring some fun and useful features to the coffee pal like: - Use your fingerprint instead your company ID Card - Coffee subscription: Get your favourite coffee automatically per mail, based on your consumption - Newsletter about your coffeeconsumption - Extend for tea, snacks and more - Connectivity through 3G for hassle-free setup. - and a lot more ;)
Now my question: Do you think such a solution could be nice for your workplace too? I know that a lot of companies are providing coffee for free, but at least in Europe a lot of my friends told me about the same problem at work.
If you are interested in this solution I would be very happy if you show your interest at the bottom of the page. I‘m curious if other people would need such a solution too.
FYI: I'm having the same problems.
Especially the click effect on buttons is really laggy (Testing on Chrome).
That's really annoying as you see it all the time.
Having a 2017 MacBook Pro, 2,9GHz I9, 16GB RAM and an Radeon with 4GB. So my Hardware shouldn't be the Problem. Maybe some Chrome/MacOSX Bug?
That may sound like an exaggeration, but it’s exactly what I see in our product.
Humans developing something already have context that agents don’t have yet. Most agents start a task with virtually no prior knowledge. And they start from zero every single time. That may improve in the future, but we’re not there yet.
Can agents get the job done? Yes. But without a thoughtfully implemented MCP server, they are awkwardly inefficient.
reply