What Is a CVE Number? Explained

A CVE number is the unique public identifier given to one specific, publicly disclosed cybersecurity vulnerability. When you see a string such as CVE-2026-125350, it is not a product name, a virus or a hacking group: it is a reference number that lets everyone — vendors, security teams, researchers and journalists — talk about the same flaw without confusion. The system is called Common Vulnerabilities and Exposures, and it is run by the non-profit MITRE Corporation under contract to the United States Cybersecurity and Infrastructure Security Agency (CISA). Its mission, as stated by CVE.org, is to identify, define and catalog publicly disclosed cybersecurity vulnerabilities, with one CVE Record for each vulnerability in the catalog.
You will have seen CVE numbers in DailyByte's cybersecurity coverage almost every day. On 15 November 2026 alone, our reports carried CVE-2026-125350, a Chrome zero-day; CVE-2026-116606, flaws in Barracuda Email Security Gateway; CVE-2026-123158, an OpenSSH flaw; CVE-2026-123039, a Check Point Gaia flaw; and CVE-2026-124194, an Outlook zero-day. This explainer answers the question behind all of those stories: what the number means, who issues it, what the score next to it means, and what you should do when one affects software you use.
What does CVE stand for?
CVE stands for Common Vulnerabilities and Exposures. Before the system existed, the same flaw could be known by half a dozen different names: the vendor's advisory number, a researcher's blog title, a scanner's plugin ID and a journalist's shorthand. That made it hard to know whether two reports described the same bug. The CVE programme, launched by MITRE in September 1999 with 321 entries, solved the naming problem by giving each publicly known flaw a single, shared identifier. Today the catalog holds hundreds of thousands of records, and the number grows every day.
A CVE Record is the published entry for one vulnerability. It carries the CVE ID, a description of the flaw, the affected products or versions, and references to the vendor advisory or researcher write-up. The record is the starting point for everything else: scanners check for it, vendors patch against it, and the United States National Vulnerability Database (NVD), run by the National Institute of Standards and Technology (NIST), adds further analysis on top, including severity scores and lists of affected products.
How to read a CVE number
Every CVE identifier follows the same format: CVE, then the year, then a sequence number — for example, CVE-2021-44228, the Log4Shell flaw in Apache Log4j disclosed in December 2021. The year is the year the identifier was reserved or the record was published; it is not necessarily the year the flaw was found, the year the software was released, or the year attacks began. The sequence number is simply the next available number in that year, and it is at least four digits long. Because so many flaws are now disclosed, identifiers routinely have five or six digits: CVE-2026-125350 is the 125,350th identifier reserved in the 2026 sequence.
Three things follow from that format. First, a high number does not mean a worse flaw; it only means the identifier was reserved later in the year. Second, two CVEs in the same year can describe completely unrelated products. Third, the number alone tells you nothing about severity, about whether a fix exists, or about whether attackers are exploiting the flaw. For all of that, you need the record and the scores attached to it.
Who hands out CVE numbers?
MITRE does not review every flaw itself. Instead, it delegates assignment to CVE Numbering Authorities, known as CNAs: vendors, open-source projects, security firms, national computer emergency response teams and bug bounty platforms, each authorised to assign identifiers within its own scope. A software vendor is usually the CNA for its own products, so when Google finds or is told about a Chrome flaw, Google's CNA assigns the CVE ID. When no vendor CNA covers a product, MITRE acts as the CNA of last resort. There are hundreds of CNAs across dozens of countries, and joining the programme costs nothing.
The process runs in stages. A researcher or vendor reports a flaw to the relevant CNA, which checks that it is a real, separately fixable vulnerability in publicly released software. The CNA reserves an identifier, and at that point the record is in a RESERVED state: the number exists, but no details are public. Once the vendor has a fix or a disclosure date, the record is PUBLISHED with its description and references. If an identifier turns out to be a duplicate, or the issue turns out not to be a vulnerability at all, the record is REJECTED. Not every flaw gets a CVE: a bug that cannot be fixed separately from another bug, or one in software that was never publicly released, generally does not qualify.
What is in a CVE record?
A published CVE record is deliberately short. It states what the flaw is, which product and versions are affected, and what an attacker could do — for example, run code remotely, read data they should not see, or crash the software. It links to the vendor advisory and to any researcher write-up. It does not, by itself, contain a severity score, a patch, or proof of exploitation. Those come from other systems that build on the CVE identifier, which is why the same CVE number appears in a vendor advisory, in the NVD, in a scanner report and in a news story, each adding its own layer of information.
That layering is the point of the system. CVE.org describes the programme's purpose as giving information technology and cybersecurity professionals a way to ensure they are discussing the same issue and to coordinate their efforts to prioritise and address vulnerabilities. A single shared name is what makes that coordination possible across companies, countries and tools that otherwise have nothing in common.
A CVE number is not a severity score
The number most often printed next to a CVE is its CVSS score. CVSS, the Common Vulnerability Scoring System, is an open standard maintained by FIRST, the Forum of Incident Response and Security Teams. It rates how severe a flaw would be if it were exploited, on a scale from 0.0 to 10.0, based on technical characteristics such as whether it can be attacked over a network, whether an attacker needs to log in, and whether the damage spreads beyond the vulnerable component. The current version is CVSS v4.0, released in November 2023, although v3.1 scores are still the most common in vendor advisories.
The severity bands are the same in both versions: 0.0 is None, 0.1 to 3.9 is Low, 4.0 to 6.9 is Medium, 7.0 to 8.9 is High, and 9.0 to 10.0 is Critical. Log4Shell, CVE-2021-44228, scored the maximum 10.0. Of the flaws reported by DailyByte on 15 November 2026, the Barracuda Email Security Gateway flaws, CVE-2026-116606, scored 9.8, which is Critical; the Chrome zero-day, CVE-2026-125350, scored 8.8, which is High; and the OpenSSH flaw, CVE-2026-123158, scored 8.1, also High. The score is usually set by the vendor or the CNA, and the NVD may publish its own score as well; the two sometimes differ.
A CVSS score measures severity, not risk. It does not say how likely the flaw is to be exploited, or how much the affected system matters to you. A Critical flaw in software you do not run is no threat to you, while a Medium flaw in a system exposed to the internet can be urgent. That is why security teams read the score alongside other signals, the most important of which is whether attackers are already using the flaw.
What is CISA's KEV list?
CISA's Known Exploited Vulnerabilities catalog, known as KEV, is the list of CVEs for which there is reliable evidence of exploitation in the wild. A flaw is added only when it has a CVE ID, there is evidence attackers are using it, and clear remediation guidance exists. United States federal agencies are required to fix KEV entries by the due date CISA sets; everyone else treats the list as a patch-first priority, whatever the CVSS score. In late October 2026 the catalog held about 1,750 entries — a tiny fraction of the hundreds of thousands of CVEs ever published.
The distinction matters. Of the flaws in DailyByte's 15 November 2026 coverage, the OpenSSH flaw CVE-2026-123158 was added to the KEV catalog on 9 November 2026, with a federal remediation deadline of 12 November 2026, and the Barracuda and Chrome flaws were also listed as known exploited. A CVE on the KEV list should be patched first, even if its score is lower than another flaw that nobody has yet exploited. Conversely, most CVEs never reach the KEV list at all, and many are never exploited.
Why are there so many CVE numbers now?
The volume of disclosures has risen sharply. An independent annual review of CVE data by researcher Jerry Gamblin counted 48,185 CVEs published in 2025, up 20.6 percent from 39,962 in 2024, taking the cumulative total since 1999 to 308,920. The rise reflects more software, more researchers looking for flaws, more vendors running disclosure programmes, and new CNAs such as the Linux kernel project, which became a CNA in 2024 and now assigns a CVE to most fixes it sends to stable branches. The flood has strained the systems that analyse CVEs: in April 2026, NIST said the NVD would fully analyse only the highest-priority CVEs going forward, starting with those on the KEV catalog, after submissions had risen 263 percent between 2020 and 2025.
The programme itself nearly stopped in April 2025, when MITRE's contract to operate it came within hours of lapsing. CISA exercised an option on the 57.8 million dollar contract, extending it for 11 months to March 2026, and the CVE Foundation was announced around the same time. The episode showed how much of the world's vulnerability tracking rests on a single programme — and why its funding became news in its own right.
What should you do when you see a CVE number?
For most readers, the practical steps are simple. First, check whether the flaw affects software or a device you actually use, and which versions are affected; the vendor advisory linked from the CVE record is the authoritative source. Second, look at the CVSS score for severity, but check the KEV list too: a KEV entry means attackers are already using the flaw, and it outranks a higher score that is not being exploited. Third, apply the vendor's patch or update, and do it sooner for internet-facing systems such as email gateways, VPNs, browsers and remote-access tools — the products that attackers can reach without being inside your network. If no patch exists yet, follow the vendor's mitigation advice, such as disabling a feature or restricting access, until one is released.
For organisations, the same logic applies at scale: inventory what you run, match it against new CVE records, prioritise KEV entries and Critical scores on exposed systems, and patch in that order. A CVE number is a tool for that triage, not a verdict on its own. It tells you exactly which flaw you are dealing with, so that the vendor's fix, the scanner's alert and the news report all line up on the same issue.
Frequently asked questions
Does having a CVE number mean a flaw is being exploited? No. A CVE number only means the flaw has been publicly disclosed and given a shared identifier. Most CVEs are never exploited. Evidence of real-world exploitation is tracked separately, notably in CISA's Known Exploited Vulnerabilities catalog.
Is a higher CVE number more dangerous? No. The sequence number only shows the order in which identifiers were reserved in a given year. Severity is shown by the CVSS score, from 0.0 to 10.0, with 9.0 to 10.0 rated Critical.
What is the difference between CVE, CVSS, NVD and KEV? CVE is the identifier and the base record for a flaw. CVSS is the scoring system that rates its severity. The NVD is the United States database, run by NIST, that adds scores, weakness types and affected-product lists to CVE records. KEV is CISA's list of the small subset of CVEs with evidence of exploitation in the wild.
Can anyone get a CVE number? Anyone can report a flaw, but only an authorised CVE Numbering Authority can assign an identifier, and only for products within its scope. Reports go to the vendor or to MITRE, which acts as the CNA of last resort.
Source: CVE.org — About the CVE Program (MITRE / CISA). Severity bands and KEV practice are described by FIRST and CISA and summarised in the CVE programme's own guidance; 2025 publication figures are from Jerry Gamblin's independent annual CVE data review. Figures were checked on 15 November 2026.