CVE Radar needs JavaScript to filter the feed.

The data is baked into this page; enable JavaScript to search, sort, and filter. The raw snapshot is also available as data.json.

Read the NVD feed the way a human would

CVE Radar is a static dashboard of newly disclosed CVEs, built from the National Vulnerability Database (NVD). Every vulnerability the NVD publishes carries a CVSS vector: a compact string like AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H that encodes how a flaw is reached and what it costs you. It is precise but unreadable at a glance. CVE Radar decodes that vector into one plain-English sentence, so you can tell a remote, no-interaction, full-impact bug from a local, low-severity one without parsing a single metric.

The whole feed filters in the browser. Sort by severity, CVSS score, or publish date; narrow by vendor or product; or type a keyword, CVE ID, or CWE class into search. The severity spectrum across the top is both a live count and a filter, and every view you build is written into the page URL, so a scoped list (say, all critical CVEs for one vendor) is a link you can paste into a ticket or a chat.

How to use it

Scan the severity spectrum to see how many critical and high CVEs are in the current snapshot.

Filter to your stack: pick a vendor or product, or search a keyword, to see only what touches you.

Read the plain-English blurb on each card to judge impact before opening the advisory.

Follow "View on NVD" or a tagged reference (patch, vendor advisory, exploit) to act.

Copy the URL to hand a teammate the exact filtered view you are looking at.

Frequently asked questions

Where does the data come from?

The National Vulnerability Database (NVD), the US government's public feed of analyzed CVEs. CVE Radar reformats the same records the NVD publishes; it never adds or edits vulnerability data.

How current is the feed?

Each build captures a point-in-time snapshot of recent CVEs. The timestamp in the masthead shows when the page was generated. For anything you plan to act on, confirm against the linked source advisory.

What do the severity levels mean?

They follow the CVSS qualitative scale: Critical (9.0 to 10.0), High (7.0 to 8.9), Medium (4.0 to 6.9), and Low (0.1 to 3.9). CVE Radar reads the published severity when present and falls back to the score band when it is missing.

How is the plain-English blurb generated?

A deterministic decoder maps each CVSS metric to words. There is no language model and no network call, so the same vector always yields the same sentence and the build stays reproducible.

Is CVE Radar a vulnerability scanner?

No. It is a reading and triage view of what has been disclosed, not a scanner of your systems. Use it to keep an eye on new CVEs and decide what deserves a closer look.