CVE Database

The AI Engine – The Chatbot, AI Framework & MCP for WordPress plugin for WordPress is vulnerable to Privilege Escalation in version 3.4.9. This is due to missing WordPress capability enforcement in the MCP OAuth bearer-token authorization path, where any valid OAuth token causes MCP access to be granted without verifying administrator privileges. This makes it possible for authenticated (Subscriber+) attackers to invoke admin-level MCP tools and escalate privileges to Administrator.

The SPIP interface_traduction_objets plugin versions prior to 2.2.2 contain an authenticated remote code execution vulnerability in the translation interface workflow. The plugin incorporates untrusted request data into a hidden form field that is rendered without SPIP output filtering. Because fields prefixed with an underscore bypass protection mechanisms and the hidden content is rendered with filtering disabled, an authenticated attacker with editor-level privileges can inject crafted content that is evaluated through SPIP's template processing chain, resulting in execution of code in the context of the web server.

ZeptoClaw is a personal AI assistant. Prior to 0.7.6, there is a Dangling Symlink Component Bypass, TOCTOU Between Validation and Use, and Hardlink Alias Bypass. This vulnerability is fixed in 0.7.6.

A critical SQL injection vulnerability in Spring AI's MariaDBFilterExpressionConverter allows attackers to bypass metadata-based access controls and execute arbitrary SQL commands. The vulnerability exists due to missing input sanitization.

Dokku is a docker-powered PaaS. Prior to 0.38.2, the openresty-vhosts plugin copies files from an app's openresty/http-includes/ git repository directory to the host and then interpolates their filenames, unescaped, into a single-quoted shell string that is later parsed by eval. A filename containing a single quote breaks the quoting and allows command substitution to execute arbitrary commands on the host as the dokku user during the app's next deploy. This vulnerability is fixed in 0.38.2.

Capgo before 12.128.2 fails to enforce limited_to_orgs and limited_to_apps constraints on subkeys provided via x-limited-key-id header in middlewareKey function. Attackers can bypass subkey scope restrictions by referencing their own subkeys, causing all downstream route handlers to use the unrestricted parent key instead of the scoped subkey.

PhreeBooks ERP 5.2.3 contains a remote code execution vulnerability in the image manager that allows authenticated attackers to upload and execute arbitrary PHP files by bypassing file extension controls. Attackers can upload malicious PHP files through the image manager endpoint and execute them to establish reverse shell connections and execute system commands.

A flaw has been found in SourceCodester Sales and Inventory System 1.0. Affected by this vulnerability is an unknown functionality of the file check_supplier_details.php of the component POST Parameter Handler. Executing a manipulation of the argument stock_name1 can lead to sql injection. The attack can be executed remotely. The exploit has been published and may be used.

In Juju from version 3.0.0 through 3.6.18, the authorization of the "secret-set" tool is not performed correctly, which allows a grantee to update the secret content, and can lead to reading or updating other secrets. When the "secret-set" tool logs an error in an exploitation attempt, the secret is still updated contrary to expectations, and the new value is visible to both the owner and the grantee.

A vulnerability was determined in Belkin F9K1015 1.00.10. This vulnerability affects the function formWlEncrypt of the file /goform/formWlEncrypt. Executing a manipulation of the argument webpage can lead to stack-based buffer overflow. The attack can be launched remotely. The exploit has been publicly disclosed and may be utilized. The vendor was contacted early about this disclosure but did not respond in any way.

OS Command Injection vulnerability in Rapid7 InsightConnect RPM Plugin on Linux allows authenticated attackers to execute arbitrary OS commands via the repo, key, or name parameters due to insufficient input sanitization in shell command construction.

In Modem, there is a possible out of bounds write due to a missing bounds check. This could lead to remote escalation of privilege, if a UE has connected to a rogue base station controlled by the attacker, with no additional execution privileges needed. User interaction is needed for exploitation. Patch ID: MOLY01088681; Issue ID: MSV-4460.

Actual is a local-first personal finance tool. Prior to version 26.4.0, any authenticated user (including `BASIC` role) can escalate to `ADMIN` on servers migrated from password authentication to OpenID Connect. Three weaknesses combine: `POST /account/change-password` has no authorization check, allowing any session to overwrite the password hash; the inactive password `auth` row is never removed on migration; and the login endpoint accepts a client-supplied `loginMethod` that bypasses the server's active auth configuration. Together these allow an attacker to set a known password and authenticate as the anonymous admin account created during the multiuser migration. The three weaknesses form a single, sequential exploit chain — none produces privilege escalation on its own. Missing authorization on POST /change-password allows overwriting a password hash, but only matters if there is an orphaned row to target. Orphaned password row persisting after migration provides the target row, but is harmless without the ability to authenticate using it. Client-controlled loginMethod: "password" allows forcing password-based auth, but is useless without a known hash established by step 1. All three must be chained in sequence to achieve the impact. No single weakness independently results in privilege escalation. The single root cause is the missing authorization check on /change-password; the other two are preconditions that make it exploitable. Version 26.4.0 contains a fix.

Type Confusion in ANGLE in Google Chrome prior to 149.0.7827.53 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: Critical)

LiteLLM through 2026-04-08 allows remote attackers to execute arbitrary code via bytecode rewriting at the /guardrails/test_custom_code URI.

A flaw was found in InstructLab. The `linux_train.py` script hardcodes `trust_remote_code=True` when loading models from HuggingFace. This allows a remote attacker to achieve arbitrary Python code execution by convincing a user to run `ilab train/download/generate` with a specially crafted malicious model from the HuggingFace Hub. This vulnerability can lead to complete system compromise.

OpenSTAManager is an open source management software for technical assistance and invoicing. Prior to version 2.10.2, multiple AJAX select handlers in OpenSTAManager are vulnerable to Time-Based Blind SQL Injection through the options[stato] GET parameter. The user-supplied value is read from $superselect['stato'] and concatenated directly into SQL WHERE clauses as a bare expression, without any sanitization, parameterization, or allowlist validation. An authenticated attacker can inject arbitrary SQL statements to extract sensitive data from the database, including usernames, password hashes, financial records, and any other information stored in the MySQL database. This issue has been patched in version 2.10.2.

Out of bounds write in ANGLE in Google Chrome prior to 149.0.7827.53 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: High)

A bug in Apache Airflow's XCom PATCH endpoint `PATCH /api/v2/xcomEntries/{key}` allowed an authenticated UI/API user with XCom write permission on a Dag to set XCom entries under reserved key names (e.g. `return_value`) that the matching POST endpoint already validated against `FORBIDDEN_XCOM_KEYS`. The endpoint also accepted serialized payload shapes the triggerer's deserializer treats as code; combined, this allowed RCE on the triggerer when the affected task next deferred. Affects deployments where untrusted users have XCom write permission on Dags that defer to the triggerer. This is a fix-bypass of CVE-2026-33858: PR #64148 added the `FORBIDDEN_XCOM_KEYS` validator only on the POST/set path; the PATCH path was not covered. Users who already upgraded for CVE-2026-33858 should additionally upgrade to `apache-airflow` 3.2.2 or later to cover the PATCH-path bypass.

OpenClaw before 2026.3.22 contains a privilege escalation vulnerability in the device.pair.approve method that allows an operator.pairing approver to approve pending device requests with broader operator scopes than the approver actually holds. Attackers can exploit insufficient scope validation to escalate privileges to operator.admin and achieve remote code execution on the Node infrastructure.

This vulnerability exists in e-Sushrut due to improper authentication logic that relies on client-side response parameters to determine authentication status. A remote attacker could exploit this vulnerability by intercepting and modifying the server response. Successful exploitation of this vulnerability could allow the attacker to bypass authentication and gain unauthorized access to user accounts on the targeted system.

e107 CMS 2.3.0 contains a remote code execution vulnerability that allows authenticated users with theme installation permissions to execute arbitrary commands by uploading malicious theme files. Attackers can upload a crafted theme package through the theme.php endpoint that deploys a web shell to the e107_themes directory, then execute system commands via the payload.php script.

Use after free in V8 in Google Chrome prior to 149.0.7827.103 allowed a remote attacker to execute arbitrary code inside a sandbox via a crafted HTML page. (Chromium security severity: High)

Remote Code Execution Vulnerability in JP1/IT Desktop Management 2 - Manager on Windows, JP1/IT Desktop Management 2 - Operations Director on Windows, Job Management Partner 1/IT Desktop Management 2 - Manager on Windows, JP1/IT Desktop Management - Manager on Windows, Job Management Partner 1/IT Desktop Management - Manager on Windows, JP1/NETM/DM Manager on Windows, JP1/NETM/DM Client on Windows, Job Management Partner 1/Software Distribution Manager on Windows, Job Management Partner 1/Software Distribution Client on Windows.This issue affects JP1/IT Desktop Management 2 - Manager: from 13-50 before 13-50-02, from 13-11 before 13-11-04, from 13-10 before 13-10-07, from 13-01 before 13-01-07, from 13-00 before 13-00-05, from 12-60 before 12-60-12, from 10-50 through 12-50-11; JP1/IT Desktop Management 2 - Operations Director: from 13-50 before 13-50-02, from 13-11 before 13-11-04, from 13-10 before 13-10-07, from 13-01 before 13-01-07, from 13-00 before 13-00-05, from 12-60 before 12-60-12, from 10-50 through 12-50-11; Job Management Partner 1/IT Desktop Management 2 - Manager: from 10-50 through 10-50-11; JP1/IT Desktop Management - Manager: from 09-50 through 10-10-16; Job Management Partner 1/IT Desktop Management - Manager: from 09-50 through 10-10-16; JP1/NETM/DM Manager: from 09-00 through 10-20-02; JP1/NETM/DM Client: from 09-00 through 10-20-02; Job Management Partner 1/Software Distribution Manager: from 09-00 through 09-51-13; Job Management Partner 1/Software Distribution Client: from 09-00 through 09-51-13.

OpenBullet2 through version 0.3.2 contains a remote code execution vulnerability that allows authenticated users to execute arbitrary commands by uploading script files (.bat.ps1.sh) through the FileProxySource proxy loading feature. Attackers can upload malicious script files as proxy sources, causing the server to execute the scripts and return output as proxy lines, resulting in arbitrary command execution on the host as the process user.

Mattermost versions 11.2.x <= 11.2.2, 10.11.x <= 10.11.10, 11.4.x <= 11.4.0, 11.3.x <= 11.3.1 fail to sanitize user-controlled post content in the mmctl commands terminal output which allows attackers to manipulate administrator terminals via crafted messages containing ANSI and OSC escape sequences that enable screen manipulation, fake prompts, and clipboard hijacking.. Mattermost Advisory ID: MMSA-2026-00599

The Photo Gallery, Sliders, Proofing and Themes – NextGEN Gallery plugin for WordPress is vulnerable to Local File Inclusion in all versions up to, and including, 4.0.3 via the 'template' parameter in gallery shortcodes. This makes it possible for authenticated attackers, with Author-level access and above, to include and execute arbitrary .php files on the server, allowing the execution of any PHP code in those files. This can be used to bypass access controls, obtain sensitive data, or achieve code execution in cases where .php file types can be uploaded and included.

A vulnerability has been found in Yi Technology YI Home Camera 2 2.1.1_20171024151200. The affected element is an unknown function of the file home/web/ipc. Such manipulation leads to hard-coded credentials. Access to the local network is required for this attack to succeed. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.

A type confusion issue was addressed with improved checks. This issue is fixed in Safari 26.5.2, iOS 26.5.2 and iPadOS 26.5.2, macOS Tahoe 26.5.2. Processing maliciously crafted web content may lead to memory corruption.

RedisTimeSeries is a time-series module for Redis. In all versions before 1.12.14 of RedisTimeSeries, the module does not properly validate serialized values processed through the Redis RESTORE command. An authenticated attacker with permission to execute RESTORE on a server with the RedisTimeSeries module loaded can supply a crafted serialized payload that triggers invalid memory access and may lead to remote code execution. A workaround is to restrict access to the RESTORE command with ACL rules. This has been patched in version 1.12.14.

The Email Address Encoder WordPress plugin before 1.0.25, email-encoder-premium WordPress plugin before 0.3.12 does not properly handle email replacement, which could allow unauthenticated users to perform Stored XSS attacks

SQL injection vulnerability in Spring AI's `CosmosDBVectorStore` allows attackers to execute arbitrary SQL queries via crafted document IDs. Affected versions: Spring AI: 1.0.0 - 1.0.5 (fixed in 1.0.6), 1.1.0 - 1.1.4 (fixed in 1.1.5)

Joomla! Component vBizz 1.0.7 contains an unrestricted file upload vulnerability that allows authenticated attackers to upload arbitrary PHP files by submitting malicious files through the profile_pic parameter. Attackers can upload PHP files via POST requests to the employee view endpoint and execute them from the uploads directory to achieve remote code execution.

In the Linux kernel, the following vulnerability has been resolved: KVM: arm64: Fix the descriptor address in __kvm_at_swap_desc() Using "(u64 __user *)hva + offset" to get the virtual addresses of S1/S2 descriptors looks really wrong, if offset is not zero. What we want to get for swapping is hva + offset, not hva + offset*8. ;-) Fix it.

A security vulnerability has been detected in UTT HiPER 1250GW up to 3.2.7-210907-180535. This issue affects the function strcpy of the file /goform/formConfigDnsFilterGlobal of the component Parameter Handler. Such manipulation of the argument GroupName leads to buffer overflow. The attack can be launched remotely. The exploit has been disclosed publicly and may be used.

A flaw was found in libinput. A local attacker who can place a specially crafted Lua bytecode file in certain system or user configuration directories can bypass security restrictions. This allows the attacker to run unauthorized code with the same permissions as the program using libinput, such as a graphical compositor. This could lead to the attacker monitoring keyboard input and sending that information to an external location.

A security vulnerability has been detected in Wavlink NU516U1 240425. Impacted is the function advance of the file /cgi-bin/wireless.cgi. Such manipulation of the argument wlan_conf/Channel/skiplist/ieee_80211h leads to os command injection. The attack may be launched remotely. The exploit has been disclosed publicly and may be used. The vendor was contacted early about this disclosure.

Contributor Privilege Escalation in Frisbii Pay <= 1.8.2 versions.

OpenClaw before 2026.3.31 contains a privilege escalation vulnerability allowing paired nodes with role=node to dispatch node.event agent requests with unrestricted gateway-side tool access. Attackers with trusted paired node credentials can escalate privileges by leveraging unrestricted agent.request dispatch to achieve remote code execution on the gateway.

The issue was addressed with improved memory handling. This issue is fixed in macOS Sequoia 15.6. Processing a maliciously crafted image may corrupt process memory.

IBM DataStage on Cloud Pak for Data 5.1.2 through 5.3.0 could allow an authenticated user to execute arbitrary commands with normal user privileges on the system due to improper validation of user supplied input through the job subroutine component.

A weakness has been identified in Edimax EW-7438RPn 1.31. This impacts the function formAccept of the file /goform/formAccept. Executing a manipulation of the argument submit-url can lead to stack-based buffer overflow. It is possible to launch the attack remotely. The exploit has been made available to the public and could be used for attacks. The vendor was contacted early about this disclosure but did not respond in any way.

A security vulnerability has been detected in Edimax EW-7438RPn 1.31. Affected is the function formRadius of the file /goform/formRadius. The manipulation of the argument submit-url leads to stack-based buffer overflow. The attack can be initiated remotely. The exploit has been disclosed publicly and may be used. The vendor was contacted early about this disclosure but did not respond in any way.

The AcyMailing – An Ultimate Newsletter Plugin and Marketing Automation Solution for WordPress plugin for WordPress is vulnerable to Missing Authorization in versions up to, and including, 10.8.2. This is due to the plugin not properly verifying that a user is authorized to perform an action. This makes it possible for authenticated attackers, with subscriber-level access and above, to modify privileged AcyMailing configuration, export subscriber secret keys, and chain these actions into administrator account takeover when a target administrator email address is known.

Hydrosystem Control System does not enforce authorization for some directories. This allows an unauthorized attacker to read all files in these directories and even execute some of them. Critically the attacker could run PHP scripts directly on the connected database.This issue was fixed in Hydrosystem Control System version 9.8.5

In versions 2.1.63 through 2.1.83 of Claude Code, the folder trust determination logic used the git worktree commondir file without validating its contents. An attacker could craft a malicious repository with a commondir file pointing to a path the victim had previously trusted, causing Claude Code to bypass its trust confirmation dialog and immediately execute hooks defined in `.claude/settings.json`. Exploitation requires the victim to clone the malicious repository and run Claude Code within it, and the attacker must know or guess a path the victim had already trusted. This issue has been fixed in version 2.1.84.

The Spectra Gutenberg Blocks – Website Builder for the Block Editor plugin for WordPress is vulnerable to Remote Code Execution in all versions up to, and including, 2.19.25. This makes it possible for authenticated attackers, with Contributor-level access and above, to execute code on the server. Exploitation requires a two-block payload embedded in post content: the first block registers a fake uagb/-prefixed block type with an attacker-specified render_callback, and the second block of the same fake type triggers invocation of that callback via call_user_func() during sequential block rendering in the same page request.

An improper certificate validation vulnerability in Ivanti Secure Access Client before 22.8R6 allows a remote unauthenticated attacker to execute arbitrary code.

Integer overflow in Blink in Google Chrome prior to 148.0.7778.96 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: Critical)

Dependency on vulnerable third-party component in GitHub Repo: zero-shot-scfoundation allows an unauthorized attacker to execute code over a network.