Minerva is an open-source technology-watch project. This page states, plainly, what data this site touches and where its content comes from.
Nothing, by default. This is a static site: no cookies, no third-party analytics, no tracking pixels, no fingerprinting. Reading Minerva leaves no trace with us.
The only personal data we ever process is your email address, and only if you deliberately subscribe to the brief. It is stored and processed by Buttondown, a US-based newsletter service, solely to send you the brief. We see your address; we never share, sell or enrich it. Every email includes a one-click unsubscribe, which also deletes your address from the list. For any further deletion request: see contact below.
Fiches are built exclusively from public repository metadata and READMEs, fetched via the official Gitee and GitHub APIs, within their published rate limits. No private data, no personal data of repository authors, no scraping.
The analytical prose (problem solved, how it works, Chinese specificity, Western equivalent) is generated by a large language model and may contain errors — fiches carry a confidence tier and may say "to be confirmed". Always verify at the source; every fiche links back to the original repository. No source code is redistributed.
If you maintain a repository covered by a fiche and want a correction — or want the fiche removed — contact us (below). We correct factual errors quickly and honor reasonable removal requests from maintainers.
Minerva's source code is Apache-2.0. Repository names, descriptions and metadata belong to their respective owners. The generated fiche text is provided as-is, for reading and sharing with attribution; it is analysis, not advice.
Contact: via the project's GitHub issues (link in the footer once the repository is public).