CVE Numbering Authorities
Up to dateWho may assign CVE IDs: organisation, country, type, scope and root.
- Records
- 567
- Data as of
- 17 Nov 2026
- Last fetched
- 17 Nov 2026
- Licence
- CVE Terms of Use
Sources
Each source is downloaded with a conditional request, checked, and only replaces the previous copy when it looks complete. A source that fails keeps its last good copy and every chart that uses it says so.
Who may assign CVE IDs: organisation, country, type, scope and root.
Every CVE record from the CVE Program: who assigned it, when it was reserved and published, the CNA's description, severity, weakness, affected products and the researchers credited; plus CISA's enrichment (SSVC, KEV) in the ADP container.
NVD's analysis of each CVE: whether and when NIST analysed it, NIST's own CVSS score, the weakness and the exact affected products (CPE).
CVEs that CISA has evidence of being exploited in the wild, with the date each was added and whether it is known to be used in ransomware campaigns.
A daily estimate, for every published CVE, of the probability it is exploited in the wild in the next 30 days; kept monthly since 2021 to see how well it predicted.
The vulnerabilities the EU Agency for Cybersecurity lists as exploited, with the date ENISA records exploitation from.
Zero-days detected being exploited in the wild, tracked by Google Project Zero, with discovery and patch dates and who reported them.
VulnCheck's known-exploited list, larger than CISA's, with the date each CVE was added. Only counts derived from it are published.
Needs an API key; not used.
Licence allows only totals to be published, not rows.
Public exploits archived by OffSec's Exploit Database, with publication dates and the CVEs they target. Exploits without a CVE reference are counted, not linked.
Ready-to-use attack modules in Rapid7's Metasploit Framework and the CVEs they reference: a strong signal that exploitation is easy.
ProjectDiscovery's community detection templates for the Nuclei scanner: a CVE with a template can be checked for at scale by anyone.
Public GitHub repositories that claim a proof-of-concept exploit for a CVE, with their creation date. A claim, not a verified exploit; some are fakes or malware.
Open-source vulnerability advisories for package ecosystems (npm, PyPI, Maven, Go, …) and Linux distributions (Debian, Ubuntu, Red Hat, Alpine, SUSE), including malicious-package reports, with CVE aliases and fixed versions.
Every advisory published by Trend Micro's Zero Day Initiative, the largest vendor-agnostic bug bounty programme: vendor, CVE, CVSS and publication date.
The Center for Threat-Informed Defense's mapping of CISA KEV entries to the ATT&CK techniques used to exploit them and the impact they enable.
Common Attack Pattern Enumeration and Classification: how weaknesses are attacked, with typical severity and likelihood, linked to CWE.
The Common Weakness Enumeration: names, hierarchy and related attack patterns of the weakness types that CVE records are classified by.
RogoLabs' published quality scores for every active CVE Numbering Authority, shown next to this census's own completeness measures for comparison.
Monthly number of papers submitted to arXiv's Cryptography and Security category, and every cs.CR paper whose title or abstract mentions CVEs, with the CVE IDs it cites.
Every main-track paper at IEEE S&P, USENIX Security, ACM CCS and NDSS since 2000, with authors, from dblp's monthly open data release.
Bug bounty reports that programmes on HackerOne chose to disclose publicly: who reported what to whom, the weakness, the severity, the bounty and any CVE IDs.