Monthly exploitation report · October 2026

Known Exploited Vulnerabilities Report — October 2026

Executive analysis of KEV additions, sensor-observed exploitation activity, CISA visibility gaps, and the vulnerabilities that drove October 2026.

160
KEVs added

151 were outside CISA KEV when added

395,497
Exploitation events

Up 84.0% from September

73.1%
Top-five concentration

Share of events tied to five vulnerabilities

1.3 days
Median CISA lead

Across 27 vulnerabilities added by Previdian first

Executive brief

The month in three decisions

01

Validate exposure against the top five

Five vulnerabilities produced 73.1% of October events. Prioritize asset discovery and remediation around PHPUnit, PHP, vite, vite-plus, react-server-dom-webpack, react-server-dom-turbopack, react-server-dom-parcel, and framework.

02

Do not rely on CISA KEV alone

Previdian added 151 exploited vulnerabilities outside CISA KEV during October; 127 were still absent at month-end.

03

Treat telemetry as directional evidence

Event volume and source breadth help identify pressure, but observations do not prove that a real-world target was compromised or represent all exploitation activity worldwide.

Exploitation activity

Activity rose sharply—and remained concentrated

Previdian sensors recorded 395,497 exploitation events in October, up from 214,910 in September. A small set of vulnerabilities drove most of the observed volume.

+84.0% month over month

Observed exploitation events

September versus October 2026

September
214,910 October
395,497

250

Vulnerabilities observed

7,323

Source IPs

27

First observed

Concentration of observed events

Top vulnerability 53.7%
Top five 73.1%
Top ten 85.0%

For CISOs, this concentration supports a focused exposure-validation and remediation sprint rather than treating every observed CVE as equally urgent.

Early Warning

17 vulnerabilities were on the autonomous watchlist first

17 of the KEVs added during October 2026 were already on the autonomous watchlist. The median lead time was 3.3 days; the longest was 22.0 days.

3.3 days

Median lead

10.6%

Share of new KEVs

CVE Publication to KEV

KEVs were added sooner after CVE publication

157 of the KEVs added during October 2026 had a CVE publication date. 66 were concurrent disclosures (CVE record and KEV on the same UTC day) and are excluded from the median. 4 were added before the CVE record was published. The median among the remaining 87 was 98 days, down from 216 days in September.

98 days

Median after publication (118 days faster than September)

42.0%

Concurrent same-day share

Most targeted

The vulnerabilities driving October activity

Ranked by sensor-observed exploitation events. Event volume shows intensity; unique source IPs help indicate breadth.

1
CVE-2017-9841

PHPUnit · PHPUnit

In CISA KEV

53.7% of monthly events

Events
212,252
Source IPs
843
3
CVE-2026-39364

Vitejs · vite, vite-plus

Outside CISA KEV at month-end

4.8% of monthly events

Events
18,926
Source IPs
611
4
CVE-2025-55182

Meta · react-server-dom-webpack, react-server-dom-turbopack, react-server-dom-parcel

In CISA KEV

3.9% of monthly events

Events
15,451
Source IPs
924
5
CVE-2026-114509

Laravel · framework

Outside CISA KEV at month-end

3.0% of monthly events

Events
11,713
Source IPs
7
7
CVE-2018-20062

NoneCms · NoneCms

In CISA KEV

2.7% of monthly events

Events
10,761
Source IPs
621
8
CVE-2022-47945

ThinkPHP · ThinkPHP Framework

Outside CISA KEV at month-end

2.6% of monthly events

Events
10,286
Source IPs
592
9
CVE-2021-41773

Apache · Apache HTTP Server

In CISA KEV

2.0% of monthly events

Events
7,717
Source IPs
967

Vendor exposure

Newly added KEVs by vendor

Plex

4 outside CISA KEV when added

4

Aspose

3 outside CISA KEV when added

3

DedeCMS

3 outside CISA KEV when added

3

Fortinet

3 outside CISA KEV when added

3

GFI USA

3 outside CISA KEV when added

3

Weakness patterns

Most common CWE classes

CWE-22 · Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')

Most common newly added weakness

34

CWE-79 · Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting')

23 newly added KEVs

23

CWE-89 · Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection')

17 newly added KEVs

17

CWE-94 · Improper Control of Generation of Code ('Code Injection')

11 newly added KEVs

11

CWE-918 · Server-Side Request Forgery (SSRF)

10 newly added KEVs

10

31.3%

Critical severity

284

Public PoC references

210

Nuclei references

248

Observed request paths

Methodology and limitations

How to interpret and cite this report

These notes define the reporting window, what sensor observations mean, and where the dataset should not be generalized.

Full Previdian methodology
Reporting period

1 October 2026 00:00 UTC through 31 October 2026 23:59 UTC.

Sensor scope

Previdian telemetry represents activity observed by the Previdian sensor network; it does not represent all exploitation activity globally.

Interpretation

An observed exploitation attempt does not by itself demonstrate that a real-world target was successfully compromised.

CISA status

Month-end status is reconstructed from stored CISA addition dates. Removals are not tracked.

Source-IP coverage

Unique source-IP counts require raw sensor rows. When only daily rollups exist, the metric is unavailable.

Historical records

Historical or backfilled KEV records cannot be identified reliably from existing fields and are not listed as a separate category.

Sensor network size

Public reports do not disclose Previdian sensor network size or per-CVE sensor counts.

Early-warning lead time

Early-warning lead time is the gap from first autonomous-watchlist listing to first Previdian KEV. Only positive gaps are counted. Listings after a KEV already existed are excluded.

CVE publication to KEV

Time from CVE publication to KEV uses UTC calendar days from the CVE publication date to the first Previdian KEV. Same-day concurrent disclosures (CVE record and KEV on the same UTC day) are counted separately and excluded from the median, as are KEVs added before the CVE record was published.

Prefer deep links to individual CVE reports and this methodology when citing sensor-observed vulnerabilities.