Vulnerability Insights.
Indexed across NVD, CISA KEV, vendor advisories, and other public vulnerability sources. Continuously updated, with CVSS, EPSS, and CWE enrichment.
Data as of March 2026Disclosure velocity.
CVEs published, modified, and added to CISA’s Known Exploited Vulnerabilities list across all tracked sources.
2025 was the all-time peak.
49,972 CVEs disclosed last year — the highest annual total ever recorded. 2026 is on pace to exceed it.
| Year | Critical | High | Medium | Low | Other | Total |
|---|---|---|---|---|---|---|
| 2026 | 1,280 | 4,512 | 5,545 | 448 | 1,420 | 13,205 |
| 2026 projected | ~6,068 | ~21,388 | ~26,285 | ~2,124 | ~6,731 | ~62,595 |
| 2025 | 4,176 | 15,555 | 25,815 | 1,269 | 3,157 | 49,972 |
| 2024 | 4,285 | 14,016 | 20,754 | 852 | 797 | 40,704 |
| 2023 | 4,633 | 10,948 | 12,815 | 421 | 2,132 | 30,949 |
| 2022 | 4,247 | 10,159 | 10,175 | 491 | 1,359 | 26,431 |
| 2021 | 2,676 | 8,567 | 8,462 | 444 | 1,801 | 21,950 |
| 2020 | 2,720 | 7,686 | 7,493 | 423 | 900 | 19,222 |
| 2019 | 2,637 | 7,206 | 7,151 | 311 | 1,633 | 18,938 |
| 2018 | 2,597 | 7,541 | 6,204 | 168 | 1,644 | 18,154 |
| 2017 | 2,111 | 6,599 | 5,697 | 235 | 3,471 | 18,113 |
| 2016 | 887 | 2,895 | 2,446 | 221 | 68 | 6,517 |
| 2015 | 48 | 2,351 | 3,506 | 589 | 101 | 6,595 |
| 2014 | 20 | 1,924 | 5,323 | 661 | 80 | 8,008 |
| 2013 | 22 | 1,732 | 2,915 | 518 | 137 | 5,324 |
| 2012 | 21 | 1,722 | 3,035 | 510 | 63 | 5,351 |
| 2011 | 5 | 1,783 | 2,101 | 261 | 22 | 4,172 |
| 2010 | 26 | 2,094 | 2,242 | 277 | 28 | 4,667 |
| 2009 | 13 | 2,734 | 2,788 | 197 | 46 | 5,778 |
| 2008 | 19 | 2,845 | 2,583 | 185 | 32 | 5,664 |
| 2007 | 21 | 3,162 | 3,101 | 232 | 80 | 6,596 |
| 2006 | 11 | 2,763 | 3,325 | 509 | 51 | 6,659 |
| 2005 | 13 | 2,047 | 2,431 | 441 | 78 | 5,010 |
| 2004 | 10 | 973 | 1,265 | 203 | 28 | 2,479 |
| 2003 | 9 | 674 | 747 | 97 | 21 | 1,548 |
| 2002 | 10 | 1,023 | 980 | 143 | 14 | 2,170 |
| 2001 | 12 | 776 | 706 | 182 | 3 | 1,679 |
| 2000 | 2 | 454 | 467 | 96 | 1 | 1,020 |
| 1999 | 3 | 422 | 356 | 113 | 29 | 923 |
| 1998 | 1 | 137 | 85 | 23 | 1 | 247 |
| 1997 | — | 145 | 87 | 20 | 1 | 253 |
| 1996 | 1 | 43 | 21 | 9 | 1 | 75 |
| 1995 | 1 | 17 | 6 | 1 | — | 25 |
| 1994 | — | 14 | 10 | 1 | 1 | 26 |
| 1993 | — | 8 | 3 | 2 | — | 13 |
| 1992 | — | 12 | — | 1 | 1 | 14 |
| 1991 | — | 11 | 4 | — | — | 15 |
| 1990 | — | 8 | 2 | 1 | — | 11 |
| 1989 | — | 2 | 1 | — | — | 3 |
| 1988 | — | 2 | — | — | — | 2 |
Which weaknesses keep showing up.
Top 10 CWE categories from 2016–2026, by year. Cross-site scripting (CWE-79) sits at the top of nearly every recent year — the ratio of XSS to other weaknesses keeps rising. Click any year to see the full ranked list.
| YEAR | CWE-79improper neutralization of input … | CWE-89improper neutralization of specia… | CWE-787out-of-bounds write | CWE-20improper input validation | CWE-125out-of-bounds read | CWE-352cross-site request forgery (csrf) | CWE-119improper restriction of operation… | CWE-200exposure of sensitive information… | CWE-862missing authorization | CWE-22improper limitation of a pathname… | TOTAL |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2026 | 1,642 | 786 | 345 | 230 | 256 | 237 | 300 | 230 | 818 | 418 | 5,262 |
| 2025 | 8,300 | 3,999 | 1,169 | 683 | 996 | 1,924 | 1,155 | 829 | 2,357 | 1,125 | 22,537 |
| 2024 | 7,146 | 2,661 | 1,996 | 682 | 1,228 | 1,387 | 294 | 800 | 1,767 | 1,066 | 19,027 |
| 2023 | 4,621 | 1,997 | 1,985 | 821 | 987 | 1,174 | 231 | 581 | 767 | 766 | 13,930 |
| 2022 | 3,216 | 1,738 | 2,271 | 751 | 864 | 694 | 339 | 486 | 512 | 743 | 11,614 |
| 2021 | 2,683 | 737 | 1,576 | 683 | 729 | 467 | 296 | 314 | 257 | 539 | 8,281 |
| 2020 | 2,167 | 463 | 1,388 | 831 | 673 | 399 | 184 | 351 | 270 | 431 | 7,157 |
| 2019 | 2,342 | 547 | 1,294 | 927 | 908 | 543 | 481 | 559 | 211 | 481 | 8,293 |
| 2018 | 2,021 | 502 | 863 | 1,281 | 768 | 453 | 1,010 | 1,075 | 67 | 564 | 8,604 |
| 2017 | 1,477 | 504 | 263 | 959 | 712 | 315 | 2,114 | 1,312 | 43 | 275 | 7,974 |
| 2016 | 472 | 94 | 175 | 526 | 90 | 81 | 1,036 | 684 | 2 | 78 | 3,238 |
| TOTAL | 36,087 | 14,028 | 13,325 | 8,374 | 8,211 | 7,674 | 7,440 | 7,221 | 7,071 | 6,486 | 115,917 |
* 2026 is partial-year data through March.
Severity, across all 338,482 CVEs.
Grouped by CVSS severity band. Critical (9.0+) and High (7.0+) together account for roughly half the catalog — a baseline reminder that the average disclosed CVE is not low-severity.
10 most recent CVEs.
Newly published vulnerability records, ordered by publication time. CVSS scores arrive after initial disclosure — recent entries may show as unscored until enrichment completes.
Improper neutralization of input during web page generation ('cross-site scripting') vulnerability in OpenText™ ZENworks Service Desk allows Cross-Site Scripting (XSS).
In Juju from version 3.0.0 through 3.6.18, when a secret owner grants permissions to a secret to a grantee, the secret owner relies exclusively on a predictable XID of the secret to verify ownership.
Deserialization of Untrusted Data vulnerability in Shinetheme Traveler allows Object Injection.This issue affects Traveler: from n/a before 3.2.8.1.
In Juju from version 3.0.0 through 3.6.18, the authorization of the "secret-set" tool is not performed correctly, which allows a grantee to update the secret content, and can lead to reading or updating other secrets.
An authorization bypass vulnerability in the Vault secrets back-end implementation of Juju versions 3.1.6 through 3.6.18 allows an authenticated unit agent to perform unauthorized updates to secret revisions.
A race condition in the secrets management subsystem of Juju versions 3.0.0 through 3.6.18 allows an authenticated unit agent to claim ownership of a newly initialized secret.
In LibreChat 0.8.1-rc2, a logged-in user obtains a JWT for both the LibreChat API and the RAG API.
LibreChat version 0.8.1-rc2 uses the same JWT secret for the user session mechanism and RAG API which compromises the service-level authentication of the RAG API.
In the Linux kernel, the following vulnerability has been resolved:
In the Linux kernel, the following vulnerability has been resolved:
Where the numbers come from.
NVD primary feed, CISA KEV catalog, vendor security advisories, and other public vulnerability sources. CVE entries reconciled across sources to a single canonical record.
CVSS for severity (v2, v3.0, v3.1, v4.0 where available), EPSS for exploitation likelihood, CWE mapping for vulnerability classification, and CISA KEV flags for known exploitation.
Records ingest continuously. Initial publications often arrive without CVSS scoring; scores fill in as CNAs and analysts complete enrichment over the following days.
Build on the live corpus through the Vulnerability APIs — CVE search, CVE detail, and the underlying scoring.
Explore more signal.
Domain Insights
391M+ domains tracked across all TLDs. Top TLDs, naming trends, and daily registration volume.
See Domain Insights API · VULNERABILITY SEARCHVulnerability Search API
Query the same 338K+ CVE corpus by vendor, product, severity, EPSS percentile, or any combination. Live data, no snapshot lag.
See API API · CVE DETAILVulnerability Detail API
Full record for any CVE: CVSS scoring across versions, EPSS exploitation likelihood, CWE classification, and CISA KEV flags.
See APIGet this data through the API.
The numbers above are point-in-time snapshots. The underlying CVE corpus is live and queryable through Deepinfo’s Vulnerability APIs — same data, with full enrichment metadata.