Injection CVEs.
Published CVEs classified as injection, ranked by exploit odds.
-
CVE-2026-93915
A missing authorization vulnerability in WatchGuard Dimension's legacy system settings configuration handler allows a low-privileged authenticated user to execute arbitrary commands with elevated privileges on the underlying OS.
-
CVE-2026-93911
A SQL injection vulnerability in WatchGuard Dimension's log collection service allows an authenticated managed device, or an attacker in possession of a valid device logging key, to achieve arbitrary code execution on the Dimension appliance.
-
CVE-2026-93912
A code injection vulnerability in WatchGuard Dimension's log notification processing allows an authenticated managed device, or an attacker in possession of a valid device logging key, to achieve arbitrary code execution on the Dimension appliance.
-
CVE-2026-130157
RepoManager is a self-hosted deb/rpm package repository. Prior to 5.13.3, an attacker who controls an upstream RPM repository mirrored by a RepoManager instance can provide a package filename through the primary.xml <location href> field; the mirror passes that filename into a shell-interpreted RPM query, allowing shell metacharacters to execute commands as the www-data service account. The target must mirror an upstream repository controlled by the attacker, and the default "Check GPG signatures" option triggers the vulnerable call. The command executes before a later signature-verification failure, so a valid signed package is not required. Fixed in 5.13.3.
-
CVE-2026-130160
RepoManager is a self-hosted deb/rpm package repository. Prior to 5.13.3, an authenticated user can supply a user-controlled, array-valued GPG key identifier to the key deletion action. Validate::string() filters HTML characters but not shell metacharacters. The user-controlled identifier reaches the delete action without shell-safe encoding. controllers/Gpg.php contains the vulnerable deletion path. Process::execute() passes the command string through proc_open() and /bin/sh -c, allowing the attacker to execute arbitrary OS commands as the web/PHP process. This can expose secrets or poison packages. Authentication is required, but the advisory makes an administrator-only restriction conditional on the intended access policy and does not establish an administrator role as a prerequisite. Fixed in 5.13.3.
-
CVE-2026-104129
pREST (PostgreSQL REST) is an API that delivers an application on top of a Postgres database. Prior to 2.4.0, the vulnerable parameter is _groupby, handled by GroupByClause in adapters/postgres/postgres.go. adapters/postgres/postgres.go uses the weaker isSafeSQLExpression check for this function-expression branch instead of the stricter single-field validator used for _select. That check validates only the outer function name against a 13-function allowlist and permits nested subqueries, trailing UNION syntax, and inner pg_ functions. controllers/crud.go concatenates the GROUP BY expression while parameterizing only WHERE values, allowing attacker-controlled SQL structure into the clause. Default configuration permits unauthenticated access; a restricted installation may require a low-privilege table reader. UNION-based extraction requires compatible column counts and types and was not verified end to end, but a delay-producing function can cause denial of service without that alignment. This issue is fixed in 2.4.0.
-
CVE-2026-103579
A CRLF injection vulnerability in WatchGuard Dimension's outgoing email server configuration allows an authenticated Dimension Administrator to execute arbitrary operating system commands on the Dimension appliance with the privileges of the local mail-handling account.
-
CVE-2026-103580
An OS command injection vulnerability in WatchGuard Dimension's package update functionality allows an authenticated Dimension Administrator to execute arbitrary operating system commands as root by submitting a specially crafted package update request.
-
CVE-2026-129221
NLTK (Natural Language Toolkit) is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. Prior to 3.10.4, nltk.classify.weka.ARFF_Formatter writes attacker-controlled class labels into the Weka ARFF @ATTRIBUTE '-label-' declaration without safely formatting the label tokens. Crafted labels can alter the generated ARFF schema or records, corrupting classifier input and causing integrity loss or processing failure. This issue is fixed in version 3.10.4.
-
CVE-2026-128138
NLTK (Natural Language Toolkit) is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. From 3.10.3 until 3.10.4, the NLTK 3.10.3 JVM hardening validates Java launcher tokens in the options channel but appends the caller-controlled cmd sequence without requiring its first token to be a main class. Supplying -jar or an @argfile through nltk.internals.java or StanfordSegmenter.java_class can therefore execute attacker-selected Java bytecode without the unsafe-JAR opt-in. This issue is fixed in version 3.10.4.
-
CVE-2026-77378
Bolt is a fully open source, adaptable platform for building and running modern websites. Prior to 6.1.4, SQL injection in ListFormatHelper allows authenticated users with upload permission to execute arbitrary SQL queries through /async/select-options and exfiltrate sensitive database contents when general/caching/list_format is enabled. This issue is fixed in version 6.1.4.
-
CVE-2026-77367
Daptin is a GraphQL/JSON-API headless CMS. Prior to 0.12.30, the generic GET /api/{entity} list endpoint accepts attacker-controlled sort values in PaginatedFindAllWithoutFilters in server/resource/resource_findallpaginated.go without validating the requested column against the table schema. The code uses those values for SELECT and ORDER BY identifier construction, so an embedded identifier quote can terminate the quoted identifier and make the remainder parse as SQL, allowing an unauthenticated attacker to extract unauthorized table data through blind or error-based behavior or repeatedly trigger query failures. This issue is fixed in version 0.12.30.
-
CVE-2026-69260
Git LFS is a Git extension for versioning large files. From 0.5.0 until 3.8.1, Git LFS does not format the SSH URL path component as a single argument when constructing remote object-transfer or authentication commands. Shell metacharacters in the SSH URL can therefore escape the intended path argument and cause arbitrary command execution when the remote Git LFS service passes received commands to a shell without adequate sanitization. Git LFS also incorrectly encodes default HTTPS URLs when an SSH URL is configured and the remote host either does not support Git LFS remote commands over SSH or supports HTTP authentication over SSH but git-lfs-authenticate does not return an explicit HTTP URL. This issue is fixed in version 3.8.1.
-
CVE-2026-133309
@scalar/typebox before 0.1.4 contains a code injection vulnerability in TypeCompiler that allows attackers to execute JavaScript by supplying crafted schema strings. Attackers can inject single-quote breakouts into string format, Kind names, property keys or Literal values to run arbitrary JavaScript in the host Node.js process during Compile() and Check().
-
CVE-2026-133323
A vulnerability in the Audio, Web, and Video Conferencing (AWV) component of Mitel MiCollab through 10.3 (10.3.0.18) could allow an unauthenticated attacker to conduct a blind command injection attack due to insufficient parameter sanitization. A successful exploit could allow an attacker to execute arbitrary commands within the context of the system. This is MTLVULN-1864.
-
CVE-2026-133303
A vulnerability in the Audio, Web, and Video Conferencing (AWV) component of Mitel MiCollab through 10.3 (10.3.0.18) could allow an unauthenticated attacker to leverage pre-staged malicious content to conduct an XML injection attack due to insufficient parameter validation. A successful exploit could allow an attacker to execute arbitrary commands within the context of the system. This is MTLVULN-1862.
-
CVE-2026-133189
A command injection vulnerability was found in OpenStack Cinder's SSH-based storage-driver support. Incomplete validation of shell metacharacters allows an authenticated user who owns a volume on an affected backend to supply crafted connection information and inject additional commands into the storage array's management interface. Commands execute with the privileges of Cinder's configured backend account. This can disclose storage configuration information and, where that account has administrative permissions, could allow changes that affect the integrity or availability of other tenants' storage.
-
CVE-2026-129412
Joomla Extension - vi-solutions - SQL injection in backend data search and export in Visforms extension 1.0.0-5.6.4, 6.0.0-6.2.1 - Search filters and form IDs in the backend data view and export were not properly sanitized, allowing SQL injection by backend users.
-
CVE-2026-129411
Joomla Extension - vi-solutions - Unauthenticated SQL injection in frontend data view filters in Visforms extension 1.0.0-5.6.4, 6.0.0-6.2.1 - Sorting, location radius and published filters of the frontend data view were not properly sanitized, allowing SQL injection, partly without authentication.
-
CVE-2026-129410
Joomla Extension - vi-solutions - Code injection via calculation formulas in Visforms extension 1.0.0-5.6.4, 6.0.0-6.2.1 - Calculation field formulas were executed using PHP eval() without validating their content, allowing backend users with field edit permissions to execute arbitrary PHP code.
-
CVE-2026-129402
Joomla Extension - vi-solutions - Unauthenticated SQL injection in reload task in Visforms extension 1.0.0-5.6.4, 6.0.0-6.2.1 - Input placeholders in SQL field statements were inserted without quoting in the public reload task, allowing unauthenticated SQL injection in read queries.
-
CVE-2026-133062
A flaw was found in Moodle's external database enrolment plugin. Insufficient sanitization of data passed to queries against the external database allows an authenticated user with course editing permissions to perform a Structured Query Language (SQL) injection attack. When this non-default plugin is enabled, an attacker can exploit this issue to view, modify, or delete sensitive records within the connected external database.
-
CVE-2026-102770
Improper neutralization of special elements in data query logic (Cypher injection) in the fulltext and vector index query generator of the sharded property database in Neo4j Enterprise Edition 2026.07.0 before 2026.10.0 allows an authenticated user holding ACCESS, CREATE INDEX and NAME MANAGEMENT privileges to inject Cypher that is accepted for execution on a property shard, in the session of any user who later queries the index, via a crafted index name containing a single quote. The impact is bounded by the privileges of the querying user.
-
CVE-2026-133194
In JetBrains Junie GitHub Action before 1.10.0 command injection was possible via the action version input
-
CVE-2026-133197
In JetBrains Junie CLI before 26.9.22 command injection was possible via file names in code review
-
CVE-2026-123891
SDDC Manager contains an SQL-injection vulnerability due to improper sanitization of user-supplied input. Affected versions: 9.1.x.x (fixed in 9.1.1.0200); 9.0.x.x (fixed in 9.0.2.0100); 5.x (fixed in 5.2.5).
-
CVE-2026-123892
SDDC Manager contains an authenticated command-injection vulnerability due to improper sanitization of user-supplied input. Affected versions: 9.1.x.x (fixed in 9.1.1.0200); 9.0.x.x (fixed in 9.0.2.0100); 5.x (fixed in 5.2.5).
-
CVE-2026-133157
flowintel contains a CSV injection vulnerability in its audit log CSV export functionality. Attacker-influenced text recorded in audit logs could be interpreted as spreadsheet formulas when an administrator opened the exported CSV file in a spreadsheet application. The vulnerable behavior allowed an attacker who could influence audit log content (for example by generating network traffic that is captured and logged) to embed formula-like strings (such as =CMD() or =HYPERLINK()) into audit entries. When the resulting CSV export was opened in a spreadsheet application, these strings could be executed as formulas, potentially leading to data exfiltration, data manipulation, or execution of further malicious actions within the spreadsheet context. Preconditions: - An attacker must be able to influence content that is recorded in flowintel audit logs. - An administrator or user must export the audit logs to CSV and open the file in a spreadsheet application that interprets formulas. Security impact: - Potential information disclosure from the spreadsheet or linked resources. - Potential integrity impact on data within the spreadsheet application. - No direct compromise of the flowintel application itself. Affected version: <4.1.0
-
CVE-2026-133100
In JetBrains TeamCity before 2026.2.2, 2026.1.6 command injection in Project Import artifact copy scripts was possible
-
CVE-2026-133092
In JetBrains TeamCity before 2026.2.2, 2026.1.6 remote code execution was possible via Git submodule URLs
-
CVE-2026-133093
In JetBrains TeamCity before 2026.2.2, 2026.1.6 code injection in parent project Kotlin DSL was possible via subproject settings
-
CVE-2026-133081
In JetBrains TeamCity before 2026.2.2, 2026.1.6 remote code execution was possible via Git VCS root URLs
-
CVE-2026-133088
In JetBrains TeamCity before 2026.2.2, 2026.1.6 remote code execution was possible via FreeMarker notification templates
-
CVE-2026-115342
Argument injection in OpenVPN version 2.4.0 through 2.6.23 and 2.7_alpha1 through 2.7.8 on Windows allows local authenticated users to bypass interactive service option validation and add arbitrary configuration options via a crafted command line.
-
CVE-2026-133023
In JetBrains TeamCity before 2026.2.1, 2026.1.5 remote code execution was possible via Kotlin DSL compiler arguments
-
CVE-2026-133026
In JetBrains TeamCity before 2026.2.1, 2026.1.5 code injection in generated Kotlin DSL was possible via matrix build parameters
-
CVE-2026-132915
In JetBrains TeamCity before 2026.2, 2026.1.4, 2025.11.8 an LDAP injection allowed bypassing login restrictions
-
CVE-2026-132909
In JetBrains TeamCity before 2026.2, 2026.1.4, 2025.11.8 remote code execution was possible via Mercurial repository contents
A list of CVEs whose flaw type is injection, sorted by how likely they are to be exploited (EPSS). Classification uses the CWE weakness id where present, and a small local model otherwise.