… what?

This sounds like the pitch of a tech bro who just heard someone use the term API but doesn’t actually know what it is

What?

There is only one reason to use an API — a developer has a use-case for it. Setting aside all of the security concerns, the existence of some API proxy is not useful in and of itself. Not only is it dead simple to set something like that up, by oneself, thus eliminating the security risks of relying on a third party, it’s just not going to gain traction if it doesn’t add something that makes it more useful than simply calling the endpoint API.

It’s cool you’re looking to try things, but perhaps coming up with a novel service would be a better use of time and effort.

I wish you luck.

What

why would i not just use the api your api hits ? are you talking about footing the bill for paid apis? then no because youre going to stop paying for those at some point.

Bro wut?

I bet every dollar I have that you don’t have the foggiest idea of wtf you’re talking about.

This place exists already; it is called GitHub.

Nope.

Honestly I would probably not use it. One if the most important things I’m looking for in an API is reliability - and a project that has no financing is really just waiting to have its plug pulled.

You might be highly motivated right now to keep it going for many years. But without a steady income it might just be a burden to keep it going at some point.

(Nothing personally obviously, I don’t know you and wish you all the best!)

Did you just get the idea for zapier?

Here is a possible case study for you. https://OpenRouter.ai.

They are doing what you think of doing, for the various LLM APIs. Three reasons why they offer added value:

1. Some people cannot access some of these APIs directly (GPT-4 in the advanced settings, Anthropic’s Claude is still close access)
2. They abstract the way these API are accessed so that the code to use different models does not have to change much.
3. For some reason, they cost the same as the original API

I’ll carry the odd opinion here and say there’s actually a way this could be useful. You have to add value to a product to make it worth your time and effort, increase adoption, and make it at least self-sustainable. Find reasons to justify why this should exist. For a start - This could save time on projects where similar data has to be loaded on a page from multiple api endpoints but it doesn’t match. - an old example, but one that I fought once - looking up the time zone of a city from one api, then the time offset from UTC from another api, and trying to relate it all together. That meant my functions had to match that data up on the client side because there were imperfect text matches.

As a second example, if you were able to cache or keep record of data from upstream endpoints that often takes a while to gather because they can’t/won’t, you might offer a performance advantage or datasets which were previously unavailable to a user without monitoring data coming from that API over an extended period of time.

There’s more you can do, but that hinges again on what I previously said, find your pitch and solve problems that the others have created and won’t fix.

It could work if you aggregated incompatible providers in the same category (such as weather) into one big aggregate API. That way, people wouldn’t need to refactor if their favorite API provider ramps up pricing or dies. But how would I trust you to keep offering the same service at a good price point instead of an established meteorological institution? Also, I think weather aggregators already exist.

APIs are already websites