Currently it only stores {email, sourceUrl} pairs in a simple object. I deliberately kept the data model minimal both for privacy and performance reasons - the

Currently it only stores {email, sourceUrl} pairs in a simple object. I deliberately kept the data model minimal both for privacy and performance reasons - the less we store, the faster it runs in the background. While it'd be technically trivial to scrape more context (DOM path, surrounding text, meta tags), I'm trying to follow the Unix philosophy of 'do one thing well' and the principle of data minimization. But I'm curious about your use case... what specific data points would be worth the privacy/performance trade-off in your opinion?