Raw vector
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XCVSS and EPSS are reproduced from their sources (NVD, FIRST EPSS). Risk Priority is our own derived reading, not an NVD score.
Summary
CVE-2026-25048 is a high-severity Uncontrolled Recursion (CWE-674) vulnerability in Mlc-Ai Xgrammar. Its CVSS base score is 8.7 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Endpoint Denial of Service (T1499); ranked in the top 49% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.
The strongest mitigations our analysis identified map to SI-10 (Information Input Validation) and SC-5 (Denial-of-service Protection) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
CVE-2026-25048 is a vulnerability in the xgrammar open-source library, which enables efficient, flexible, and portable structured generation. In versions prior to 0.1.32, processing multi-level nested syntax triggers a segmentation fault resulting in a core dump. The issue carries a CVSS 3.1 base score of 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) and maps to CWE-674 (Uncontrolled Recursion).
Attackers require only network access to exploit this vulnerability, with low attack complexity, no privileges, and no user interaction needed. Successful exploitation causes high-impact availability disruption through application crashes via segmentation faults, enabling denial-of-service attacks against systems using vulnerable xgrammar instances.
The vulnerability has been patched in xgrammar version 0.1.32. Security practitioners should upgrade to this version or later, as detailed in the project's GitHub release notes (https://github.com/mlc-ai/xgrammar/releases/tag/v0.1.32) and security advisory (https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-7rgv-gqhr-fxg3).
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-9830
Vulnerability Data
xgrammar is an open-source library for efficient, flexible, and portable structured generation. Prior to version 0.1.32, the multi-level nested syntax caused a segmentation fault (core dumped). This issue has been patched in version 0.1.32.
- CWE(s)
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Mitigating Controls (NIST 800-53 r5) AI
Input validation can reject or constrain data that would otherwise drive unbounded recursive calls.
DoS protection mechanisms limit the resource-exhaustion impact of uncontrolled recursion without eliminating the flaw.
System monitoring can observe anomalous resource consumption that signals runaway recursion after it begins.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Secure SDLC practices directly prevent coding errors such as missing recursion limits or termination conditions.
Runtime monitoring of compute resources can detect excessive consumption caused by uncontrolled recursion.
Vulnerability identification processes can discover and record uncontrolled recursion flaws before deployment.
Capacity monitoring and resource provisioning can absorb or limit the impact of runaway recursion.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Security testing in development can detect excessive recursion via static analysis or fuzzing.
Secure development life cycle requires controls that prevent uncontrolled recursion through design and code review.
Application security requirements can mandate recursion limits or stack-depth checks.
Secure system architecture principles include resource-management and input-validation rules that limit recursion.
Secure coding standards directly prohibit or constrain recursive constructs that could exhaust stack or memory.
Capacity management includes monitoring and limits that mitigate resource exhaustion from runaway recursion.