Bot for Kraken
Run a simple simulated strategy locally, but using real data from the exchange, do:
python -m boken
Disclaimer: Currently it only runs basic example requests while core functionality is being developed.
To use this library via a jupyter notebook to quickly get a nice visual feedback, follow instructions in JupyterLab.md Then open a notebook and enter:
TODO
To setup your keys to access your account, do:
TODO
Note : on first start, the ledger for the bot trade is created as a sqlite database. There is one database per strategy, and it holds all of the state the bot could have and needs to persist between run. To work interactively with the sqlite DB, do:
TODO
To run a few commands manually for kraken, do:
TODO
Note the trades made like this will be affected to a specific ledger named "cli"
To run a simple strategy, for real, on kraken, do:
TODO
We follow gitflow branching strategy. Any question/concern, do not hesitate to post an Issue. Any proposal for improvement, do not hesitate to post a Pull Request. We will try to prioritize addressing issues and merging PR via milestones.
Disclaimer : We are currently setting up the basis of the code, so this can be messy at the moment...
We will use the tools that makes it convenient to code interactively in python. Toolchain:
- pipenv
- pytest
- vcrpy
- mypy
- black
- jupyter
- sphynx
Feel free to propose your favorite tool and argument why it would help to bring in a better interactivity when working with an autonomous software.
To run the tests, using prerecorded server response, do:
pytest
This includes typing and doctests.
To get meaningful server data to test again later, we use vcrpy. to run the tests and record, do:
pytest --record-mode=all