I'm working on BankBotsBank (https://github.com/bankbotsbank/bankbotsbank), an open effort to develop a collection of open-source bots that provide API access to accounts on any bank.
I developed the first bot, which can be used as a sample, with JavaScript (ES6, and ES7's async/await) and Nightmare (based on Electron, so on Chromium).
It's awesome because free API access to this data for any bank in the world may enable new and interesting fintech apps (actually, for me this project is a stepping stone towards another project that I plan to build on top of it), and I think that the open-source approach is the most effective way to achieve that (i.e. free and comprehensive API access to accounts data on any bank of the world).
I've actually been thinking about such an effort because I was sick of Mint being more geared at selling me loans than giving me the tools that others and I have requested, and none of the other services that automatically pull in bank data are much better.
Have you thought of making some sort of framework that can handle security and aggregate the data, and have developers plug in the bank specific bots?
My idea is to keep it simple: just a collection of bots that accept any required credentials and return accounts data in a common JSON format. Storing credentials and aggregated data is not in the scope of this project. If that's needed (not all applications may need it), it has to be handled by the client app.
I'm currently experimenting with classifiers to make semi-automated expenses classification. My bank don't offer API access (of course. Because they still live in the dark ages), but they do have an "usable" CSV export. What would be really awesome is a raspberry pi running a bot once a day to grab a fresh CSV, then parse, classify, aggregate and historize it.
The funny thing is, they do have mobile/tablet applications and other "new" stuff, but they just completely sucks.
I developed the first bot, which can be used as a sample, with JavaScript (ES6, and ES7's async/await) and Nightmare (based on Electron, so on Chromium).
It's awesome because free API access to this data for any bank in the world may enable new and interesting fintech apps (actually, for me this project is a stepping stone towards another project that I plan to build on top of it), and I think that the open-source approach is the most effective way to achieve that (i.e. free and comprehensive API access to accounts data on any bank of the world).