CVE-2026-55962 is a TLS 1.3 post-handshake authentication flaw in the wolfSSL embedded TLS library: a server that requests a client certificate after the handshake could accept the client's Finished message without a Certificate and CertificateVerify. NVD scores it 6.5, only servers built with --enable-postauth are affected, and wolfSSL 5.9.2 fixes it. OpenWrt lists it among its 5.9.2 update fixes.
Vulnerability summary
CVSS base score
6.5
Medium severity
EPSS score (FIRST)
0.2%
Probability of exploitation in the wild in the next 30 days
CISA KEV catalog
No
Not listed
Published
Vulnerability ID: CVE-2026-55962 · Source: NVD
Affected components
| Component | CPE | Version range |
|---|---|---|
| wolfSSL (TLS 1.3 post-handshake authentication, server side) | cpe:2.3:a:wolfssl:wolfssl:*:*:*:*:*:*:*:* | NVD: 5.5.4 up to (excluding) 5.9.2; only builds with WOLFSSL_POST_HANDSHAKE_AUTH (--enable-postauth); fixed in 5.9.2 (PR 10702) |
| OpenWrt libwolfssl package | — | 5.7.6 or 5.9.1 in OpenWrt 24.10.0-24.10.8, 5.8.4 or 5.9.1 in 25.12.0-25.12.5 (stock build has no --enable-postauth); 5.9.4 in 25.12.6 |
What does this mean for affected devices?
CVE-2026-55962 is an improper authentication flaw (CWE-287, per wolfSSL as CNA) in wolfSSL's TLS 1.3 server code. The 5.9.2 section of the wolfSSL ChangeLog (23 June 2026) explains that the exemption allowing an empty or absent peer certificate "was only intended for the initial handshake, but it was also being applied while a post-handshake CertificateRequest was still outstanding". It "only affects TLS 1.3 servers built with post-handshake authentication support (WOLFSSL_POST_HANDSHAKE_AUTH / --enable-postauth, included in --enable-all)" that enable WOLFSSL_VERIFY_POST_HANDSHAKE and call wolfSSL_request_certificate(); clients and servers that do not use post-handshake authentication are unaffected. NVIDIA Project Vanessa reported it, and pull request 10702 fixes it.
NVD gives it a Primary CVSS v3.1 score of 6.5 with vector CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:N and covers wolfSSL 5.5.4 up to (excluding) 5.9.2; wolfSSL's CVSS v4.0 score is 6.0 and the wolfSSL vulnerabilities page rates it Medium (NVD). On a device the pattern to look for is a TLS 1.3 server, such as a management interface, that asks for a client certificate only after the connection is up.
OpenWrt's 5.9.2 update commit lists CVE-2026-55962 (Med) among 32 fixed CVEs and notes that "many of them apply to the configuration built here". This one needs --enable-postauth, which the stock OpenWrt wolfSSL Makefile does not pass and wolfSSL's configure.ac leaves disabled by default, so a stock OpenWrt libwolfssl is outside the condition even where its version is in range (5.9.1 in 24.10.7, 24.10.8 and 25.12.3 to 25.12.5). Firmware built with --enable-all includes post-handshake authentication.
Debian fixed it in bookworm with 5.5.4-2+deb12u3 (DLA-4683-1), in trixie with 5.7.2-0.1+deb13u2 and in sid with 5.9.2-1 (Debian security tracker).
CVE-2026-55962 is not in the CISA KEV catalog as of 2026-11-16 (catalog version 2026.11.13), and FIRST EPSS gives it 0.0024, the 14.5th percentile, in the 2026-11-15 dataset (CISA KEV, FIRST EPSS).
Check the build for WOLFSSL_POST_HANDSHAKE_AUTH and the application for wolfSSL_request_certificate(); if both are present, move to 5.9.2 or later. Our Mbed TLS and wolfSSL firmware guide covers reading the build configuration, and our authentication bypass fix guide covers the class. Other wolfSSL authentication flaws are CVE-2026-93302 and CVE-2026-5194.
Evidence
- Post-handshake exemption also applied while a post-handshake CertificateRequest was outstanding; only TLS 1.3 servers with WOLFSSL_POST_HANDSHAKE_AUTH / --enable-postauth (included in --enable-all); reported by NVIDIA Project Vanessa; fixed in PR 10702, release 5.9.2 (23 June 2026). — wolfSSL ChangeLog (5.9.2)
- Medium rating, fixed in 5.9.2. — wolfSSL security vulnerabilities page
- NVD Primary CVSS v3.1 6.5 (AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:N); CNA CVSS v4.0 6.0; CPE range 5.5.4 to before 5.9.2. — NVD
- OpenWrt 5.9.2 update lists CVE-2026-55962 (Med) and says many of the CVEs apply to its configuration. — OpenWrt commit 21945f87
- OpenWrt configure arguments contain no --enable-postauth. — OpenWrt wolfSSL Makefile (v24.10.8)
- --enable-postauth default disabled. — wolfSSL configure.ac (v5.9.2-stable)
- Debian: bookworm 5.5.4-2+deb12u3 (DLA-4683-1), trixie 5.7.2-0.1+deb13u2, sid 5.9.2-1 fixed. — Debian security tracker
- Not in the CISA KEV catalog as of 2026-11-16 (catalog version 2026.11.13). — CISA KEV
- EPSS 0.00235 (14.5th percentile), dataset 2026-11-15, read 2026-11-16. — FIRST EPSS

Related advisories
Sources
Frequently asked questions
Which devices are affected by CVE-2026-55962?
Only devices running a wolfSSL TLS 1.3 server built with WOLFSSL_POST_HANDSHAKE_AUTH (--enable-postauth, also part of --enable-all) that enable WOLFSSL_VERIFY_POST_HANDSHAKE and request a client certificate after the handshake. Clients are unaffected (per the wolfSSL ChangeLog).
Is a stock OpenWrt build affected by CVE-2026-55962?
OpenWrt listed the CVE in its 5.9.2 update, but its stock Makefile does not pass --enable-postauth and wolfSSL leaves the option disabled by default, so the stock libwolfssl does not meet the condition; 25.12.6 ships 5.9.4 (per OpenWrt and wolfSSL configure.ac).
Which version fixes CVE-2026-55962?
wolfSSL 5.9.2, through pull request 10702; NVD covers 5.5.4 up to (excluding) 5.9.2. Debian fixed it in bookworm (DLA-4683-1), trixie and sid (per the wolfSSL ChangeLog, NVD and Debian).
Is CVE-2026-55962 exploited in the wild?
It is not in the CISA KEV catalog as of 2026-11-16 (catalog version 2026.11.13), and FIRST EPSS gives it 0.0024, the 14.5th percentile, in the 2026-11-15 dataset (per CISA KEV and FIRST EPSS).
Questions about this page?