Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessCraft CMS message with 4-byte aes-ICVlen
-
DeliveryDeliver malformed message to target via email or file
-
ExploitTarget invokes gpgsm to decrypt or verify
-
ExecutionParser accepts truncated ICV without error
-
PersistAES-GCM integrity check bypassed
-
ImpactForged or tampered ciphertext accepted as valid
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V2 - Validation and Business Logic Input size validation
|
| Exploitation Conditions | Exploitation requires that the target runs GnuPG through 2.5.20 and uses the gpgsm component to process externally-supplied CMS-formatted data containing AES-GCM encryption. The CVSS AV:L metric indicates the malicious CMS content must be processed locally - while the crafted message can arrive via email or file transfer, gpgsm must be invoked against it on the local system. AC:H further constrains exploitation, requiring precise crafting of the CMS structure to manipulate the aes-ICVlen field to 4 bytes while maintaining a parseable envelope. No authentication on the target system is required per PR:N. Systems that do not use gpgsm, do not process S/MIME or CMS content, or do not accept CMS input from untrusted sources are not meaningfully exposed. The 'Information Disclosure' tag applied in source intelligence conflicts with the CVSS C:N metric; confidentiality impact is not confirmed by any available data and should not be assumed. |
| Impact | The CVSS 3.1 base score of 2.9 (Low) with vector AV:L/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N places this at the lowest practical threat tier. No EPSS data was provided in the available intelligence, so exploitation probability cannot be quantified numerically. The vulnerability is not listed in the CISA KEV catalog, and no public POC is referenced in the available data. The 'Information Disclosure' tag applied to this CVE appears inconsistent with the CVSS C:N metric - this discrepancy should be verified with the vendor, as the CVE description does not clearly support a confidentiality impact and may reflect erroneous tagging. The primary real-world risk is cryptographic: accepting a 4-byte authentication tag instead of 12 bytes drastically reduces the security margin of AES-GCM integrity verification from 2^96 to 2^32 combinations, potentially enabling forged or tampered CMS messages to pass verification. Despite this cryptographic concern, AV:L and AC:H substantially limit practical exploitability, making this a low operational priority unless the environment processes untrusted CMS/S/MIME content at scale. |
| Exploit ScenarioDerived | An attacker crafts a malicious CMS-formatted message containing AES-GCM ciphertext with a deliberately shortened 4-byte ICV field rather than the required 12 bytes, then delivers it to a target who processes it with gpgsm (for example, via an S/MIME email). The parser accepts the truncated ICV without raising a validation error, potentially allowing forged or attacker-modified ciphertext to pass AES-GCM integrity verification undetected. No public POC has been identified; successful exploitation would require precise knowledge of CMS encoding structure and GnuPG's internal parsing behavior. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Jun 23, 2026
GnuPG's gpgsm component through version 2.5.20 improperly validates AES-GCM authentication tag length during CMS parsing, accepting a 4-byte ICV where the cryptographic standard mandates 12 bytes. This validation failure means gpgsm will process CMS-formatted messages with a truncated integrity check value, undermining the authentication guarantee that AES-GCM is specifically designed to provide. No public exploit has been identified at time of analysis; the low CVSS score of 2.9 reflects constrained attack conditions, though the related CVE-2026-34182 warrants cross-referencing as it may share the same code path.
RemediationAI
The primary remediation is to upgrade GnuPG to a version released after 2.5.20 that addresses this ICV length validation flaw; consult the official GnuPG download page at https://www.gnupg.org/download/ for the latest release, as no specific patched version number was confirmed in the available intelligence data. If an immediate upgrade is not feasible, a targeted compensating control is to restrict gpgsm from processing CMS-formatted AES-GCM content sourced from untrusted external parties - for example, disable automatic S/MIME decryption for inbound email from external senders, accepting the trade-off of requiring manual processing or alternative tooling. Additionally, CVE-2026-34182 is explicitly noted as related and should be investigated and remediated concurrently, as it may share the same vulnerable code path.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
CMS (Cryptographic Message Syntax, RFC 5652) is the format underlying S/MIME encrypted email and PKCS#7 structures. AES-GCM (AES in Galois/Counter Mode) is an authenticated encryption scheme whose security depends critically on the integrity check value (ICV), also known as the authentication tag. Per NIST SP 800-38D and the CMS AES-GCM extension RFCs, the ICV must be 12 bytes (96 bits) to ensure negligible forgery probability. The affected component, gpgsm, is GnuPG's implementation for S/MIME and CMS operations. CPE cpe:2.3:a:gnupg:gnupg:*:*:*:*:*:*:*:* covers all GnuPG releases through 2.5.20. CWE-1284 (Improper Validation of Specified Quantity in Input) precisely characterizes the root cause: the parser accepts a user-controlled length field (aes-ICVlen) without enforcing the required 12-byte minimum, allowing a 4-byte value to pass unchallenged.
Know when Gnupg gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:L/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:N
CMS messages are delivered over network channels such as email (AV:N); crafting a valid malformed ICV requires expertise (AC:H); a user must trigger gpgsm processing (UI:R); no confidentiality impact is supported by the description.
Primary rating from Vendor (mitre).
CVSS VectorVendor: mitre
DescriptionCVE.org
CMS (Cryptographic Message Syntax) parsing in gpgsm in GnuPG through 2.5.20 mishandles the CMS format for AES-GCM because aes-ICVlen is supposed to be 12 bytes but 4 bytes is accepted. NOTE: this is related to CVE-2026-34182.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
Remote code execution and denial of service in GnuPG before 2.5.17 stem from a stack-based buffer overflow in gpg-agent
GnuPG version 2.1.12 - 2.2.11 contains a Cross ite Request Forgery (CSRF) vulnerability in dirmngr that can result in At
cipher/elgamal.c in Libgcrypt through 1.8.2, when used to encrypt messages directly, improperly encodes plaintexts, whic
Stack-based buffer overflow in GnuPG's tpm2daemon (versions before 2.5.17) allows a local attacker to corrupt the daemon
GnuPG 2.2.21 and 2.2.22 (and Gpg4win 3.1.12) has an array overflow, leading to a crash or possibly unspecified other imp
Interaction between the sks-keyserver code through 1.2.0 of the SKS keyserver network, and GnuPG through 2.2.16, makes i
GnuPG through 2.3.6, in unusual situations where an attacker possesses any secret-key information from a victim's keyrin
The read_block function in g10/import.c in GnuPG 1.4.x before 1.4.13 and 2.0.x through 2.0.19, when importing a key, all
The ElGamal implementation in Libgcrypt before 1.9.4 allows plaintext recovery because, during interaction between two c
Libksba before 1.6.3 is prone to an integer overflow vulnerability in the CRL signature parser. Rated critical severity
Integer underflow in the ksba_oid_to_str function in Libksba before 1.3.2, as used in GnuPG, allows remote attackers to
Libgcrypt before 1.7.10 and 1.8.x before 1.8.3 allows a memory-cache side-channel attack on ECDSA signatures that can be
EUVD-2026-38550
GHSA-m6x2-4hhh-669j
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB