THU-MIG Yoloe code injection vulnerability

THU-MIG Yoloe has a code injection vulnerability. This vulnerability stems from improper handling of parameters in the eval function of the ultralytics/nn/autobackend.py file within the AutoBackend Model Loading component. This improper handling leads to the of instructions in the dynamically...

Eval Injection

Overview sglang is a SGLang is a fast serving framework for large language models and vision language models. Affected versions of this package are vulnerable to Eval Injection via evaluatefilter, which passes attacker-controlled input to eval. A remote attacker can escape the eval sandbox by...

CVE-2026-111391

SGLang 0.5.10 through 0.5.22 contains a code injection vulnerability in evaluatefilter that allows remote attackers to execute Python code by escaping the eval sandbox through re.search globals. Attackers can send a crafted filter expression to the unauthenticated /dumper/configure endpoint when...

CVE-2026-111391 SGLang 0.5.10 through 0.5.22 Code Injection via Debug Dumper Filter

SGLang 0.5.10 through 0.5.22 contains a code injection vulnerability in evaluatefilter that allows remote attackers to execute Python code by escaping the eval sandbox through re.search globals. Attackers can send a crafted filter expression to the unauthenticated /dumper/configure endpoint when...

EUVD-2026-102845

SGLang 0.5.10 through 0.5.22 contains a code injection vulnerability in evaluatefilter that allows remote attackers to execute Python code by escaping the eval sandbox through re.search globals. Attackers can send a crafted filter expression to the unauthenticated /dumper/configure endpoint when...

CVE-2026-111237 Ultralytics FastSAM AutoBackend Model Loading autobackend.py session.get_modelmeta.custom_metadata_map eval injection

A weakness has been identified in Ultralytics FastSAM up to b4ed20c2fed75eadc5aa7d8b09fedd137b873b52. This affects the function session.getmodelmeta.custommetadatamap of the file ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. This manipulation of the argument...

CVE-2026-111237 Ultralytics FastSAM AutoBackend Model Loading autobackend.py session.get_modelmeta.custom_metadata_map eval injection

A weakness has been identified in Ultralytics FastSAM up to b4ed20c2fed75eadc5aa7d8b09fedd137b873b52. This affects the function session.getmodelmeta.custommetadatamap of the file ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. This manipulation of the argument...

CVE-2026-111237 Ultralytics FastSAM AutoBackend Model Loading autobackend.py session.get_modelmeta.custom_metadata_map eval injection

A weakness has been identified in Ultralytics FastSAM up to b4ed20c2fed75eadc5aa7d8b09fedd137b873b52. This affects the function session.getmodelmeta.custommetadatamap of the file ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. This manipulation of the argument...

CVE-2026-111237

A vulnerability exists in Ultralytics FastSAM (up to commit b4ed20c2fed75eadc5aa7d8b09fedd137b873b52) within the AutoBackend Model Loading component. Specifically, the function session.get_modelmeta.custom_metadata_map in ultralytics/nn/autobackend.py is susceptible to an eval injection (CWE-95, ...

CVE-2026-111212 ImprintLab Medical-SAM-Adapter AutoBackend Model Loading autobackend.py session.get_modelmeta.custom_metadata_map eval injection

A security flaw has been discovered in ImprintLab Medical-SAM-Adapter 0.1.0-alpha/0.1.0. Affected by this issue is the function session.getmodelmeta.custommetadatamap of the file models/MobileSAMv2/ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. The manipulation of the...

CVE-2026-111212

A security flaw exists in ImprintLab Medical-SAM-Adapter versions 0.1.0-alpha and 0.1.0 . The vulnerability is located within the AutoBackend Model Loading component, specifically in the session.get_modelmeta.custom_metadata_map function of the file models/MobileSAMv2/ultralytics/nn/autobackend.p...

CVE-2026-111212 ImprintLab Medical-SAM-Adapter AutoBackend Model Loading autobackend.py session.get_modelmeta.custom_metadata_map eval injection

A security flaw has been discovered in ImprintLab Medical-SAM-Adapter 0.1.0-alpha/0.1.0. Affected by this issue is the function session.getmodelmeta.custommetadatamap of the file models/MobileSAMv2/ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. The manipulation of the...

CVE-2026-111212 ImprintLab Medical-SAM-Adapter AutoBackend Model Loading autobackend.py session.get_modelmeta.custom_metadata_map eval injection

A security flaw has been discovered in ImprintLab Medical-SAM-Adapter 0.1.0-alpha/0.1.0. Affected by this issue is the function session.getmodelmeta.custommetadatamap of the file models/MobileSAMv2/ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. The manipulation of the...

Astra Linux – Vulnerabilities in Linux 5.10, Linux 5.15

This vulnerability allows local attackers to disclose sensitive information on affected installations of the Linux Kernel 6.0-rc2. An attacker must first gain the ability to execute high-privileged code on the target system in order to exploit this vulnerability. The specific flaw resides within...

Astra Linux – Vulnerability in libio-compress-perl

IO::Compress versions before 2.220 for Perl can execute arbitrary code in File::GlobMapper via an attacker-controlled output glob. The parseOutputGlob function wraps the output glob string provided by the caller within double quotes and stores it in the parser state. Then, the getFiles function...

Astra Linux – Vulnerability in libmodule-scandeps-perl

Qualys discovered that if unsanitized input was used with the Modules::ScanDeps library, before version 1.36, a local attacker could potentially execute arbitrary shell commands by opening a “pesky pipe” e.g., passing “commands|” as a filename or by passing arbitrary strings to the eval function...

PYSEC-2026-4232 Langflow: Prompt injection in Langflow Smart Transform can lead to code execution

Summary Langflow versions 1.3.0 through 1.10.2 contain a code-injection vulnerability in the Smart Transform LambdaFilterComponent component. Smart Transform places flow-author instructions and a preview of its input data into a prompt asking an LLM to generate a Python lambda. It then extracts a...

Langflow: Prompt injection in Langflow Smart Transform can lead to code execution

SummaryLangflow versions 1.3.0 through 1.10.2 contain a code-injection vulnerability in the Smart Transform LambdaFilterComponent component.Smart Transform places flow-author instructions and a preview of its input data into a prompt asking an LLM to generate a Python lambda. It then extracts a...

freeIPA: idm: freeipa: idp-add eval() reachable before authorization check allows environment disclosure and denial of service

A flaw was found in FreeIPA's idp-add command, where insufficiently validated --organization/--base-url input reaches a constrained eval call before the corresponding LDAP access control check is enforced. This allows any authenticated IPA principal, regardless of privilege level, to enumerate an...

freeIPA: idm: freeipa: idp-add eval() reachable before authorization check allows environment disclosure and denial of service

A flaw was found in FreeIPA's idp-add command, where insufficiently validated --organization/--base-url input reaches a constrained eval call before the corresponding LDAP access control check is enforced. This allows any authenticated IPA principal, regardless of privilege level, to enumerate an...