I am working on a project that will make it easier to implement value based algorithms in Zipline. The code is @ https://github.com/andrewkittredge/financial_fundamentals, you can pip install it from https://pypi.python.org/pypi/FinancialFundamentals/.
FinancialFundamentals has a couple components. It scrapes the SEC's edgar system to find xbrl filings. Then it parses out accounting metrics, currently it only has EarningsPerShare. It can cache the values in either SQLite or mongodb, SQLite caching works with no configuration.
It can also cache price data downloaded from yahoo so you don't have to hit yahoo every time you run your algorithm. There is a tongue-in-cheek example @ https://github.com/andrewkittredge/financial_fundamentals/blob/master/examples/warren_buffet.py.
I am actively developing the project, I would love feedback and bug-reports. It doesn't integrate with quantopian as of now but I might put up something so you could get accounting metrics using quantopians csv importing feature.