Check conda-forge package versions against their sources to determine if they are out of date. An example:

$ feedstockrot cherrypy matplotlib flake8 docker-py netcdf4 redis-py 
Up-to-date:
- netcdf4
- redis-py
- matplotlib
Upgradeable:
- cherrypy: 6.1.1 -> 7.1.0
- docker-py: 1.8.1 -> 1.9.0
- flake8: 2.6.2 -> 3.0.4

conda install --channel conda-forge feedstockrot

Or, in its own environment:

conda create --name feedstockrot feedstockrot
source activate feedstockrot

To start, just execute and specify which packages to check:

feedstockrot cherrypy matplotlib flake8 netcdf4 redis-py

To check all of your feedstocks, export a GitHub token and run:

export FEEDSTOCKROT_GITHUB_TOKEN=your-github-token
feedstockrot --github

Setup a conda environment and install dependencies:

conda create -name feedstockrot-dev --file requirements.txt python=3.5
source activate feedstockrot-dev
conda install --file requirements-dev.txt

Optionally, install this tool in editable mode. This isn't required, as main.py can be executed directly.

pip install --editable .

Brought to you by Axiom Data Science.