Methodology
What DayZeroHub is, where the advisories come from and how we write them.
About the name
The name DayZeroHub does not mean every advisory is a zero-day vulnerability. We publish vulnerability advisories from many sources and only flag exploitation as known when an official source explicitly backs it (for example, CISA's KEV catalog). We do not label a vulnerability as zero-day without explicit evidence.
Sources
We favor official APIs, JSON and feeds. We keep the URL, identifier, dates and a hash of the original content so every data point is traceable. We do not republish third-party articles in full: we publish our own summaries and link to the source.
Dates
We distinguish the original publication date, the date we ingested it and the date of the last update. An import of older data is never presented as a vulnerability published today.
How an advisory is built
Collection, normalization, deduplication by identifiers (never by title similarity), writing in English and Spanish, validation against the structured source data, and publication.
Use of artificial intelligence
Summaries are written with AI assistance from the source data. Validation checks identifiers, versions, scores, dates, exploitation status and references against the structured sources; when there is a conflict or missing evidence, the advisory is not published automatically. The AI does not produce exploit code or instructions for exploiting vulnerabilities. Every advisory links to its sources.
What it does not do
DayZeroHub does not scan networks or servers, does not audit sites, does not exploit vulnerabilities and does not know what software you run: choosing a product to follow in your alerts does not imply that you use it.
Limits
Data depends on the sources and may change or be corrected. Always verify information at the original source before acting.