CVE-2026-116275
Received Received - Intake

Code Injection in AllenNLP via Registrable Configuration

Publication date: 2026-10-26

Last updated on: 2026-10-26

Assigner: VulDB

Description
A vulnerability was found in AllenAI AllenNLP up to 2.10.1. This impacts the function Registrable.resolve_class_name of the file allennlp/common/registrable.py of the component Registrable Configuration Machinery. Performing a manipulation of the argument Type results in code injection. The attack may be initiated remotely. The exploit has been made public and could be used. The vendor was contacted early about this disclosure but did not respond in any way. This vulnerability only affects products that are no longer supported by the maintainer.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-10-26
Last Modified
2026-10-26
Generated
2026-10-26
AI Q&A
2026-10-26
EPSS Evaluated
N/A
NVD
CVE-2026-116275
EUVD
EUVD-2026-109790

Affected Vendors & Products

Showing 2 associated CPEs
Vendor Product Version / Range
AllenAI AllenNLP 2.10.0
AllenAI AllenNLP 2.10.1

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-74 The product constructs all or part of a command, data structure, or record using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify how it is parsed or interpreted when it is sent to a downstream component.
CWE-94 The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

CVE-2026-116275 is a code injection vulnerability in AllenNLP version 2.10.1, an end-of-life natural language processing framework. It occurs in the Registrable.resolve_class_name function due to unsafe reflective instantiation in the FromParams/Registrable configuration system. The flaw allows attackers to execute arbitrary remote code by manipulating the Type argument in a model archive's config.json file.

Detection Guidance

Detecting this vulnerability requires checking for AllenNLP version 2.10.1 or earlier and inspecting model archives for suspicious config.json files. Look for configurations using unregistered classes like subprocess.Popen in dataset_reader.type or similar fields. Check system logs for unexpected command executions during AllenNLP operations.

Impact Analysis

This vulnerability enables attackers to execute arbitrary code on your system with the privileges of the user running AllenNLP. Attackers can embed malicious payloads in model archives, leading to file system compromise, data exfiltration, or other destructive actions when the archive is loaded via commands like allennlp predict or allennlp train.

Mitigation Strategies

Immediately upgrade to a supported AllenNLP version if available, or stop using AllenNLP 2.10.1 entirely. Avoid loading untrusted model archives. Run AllenNLP in isolated, unprivileged environments with minimal permissions. Monitor for suspicious activities in system logs.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-116275. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart