Claude Code is an agentic coding tool. Prior to version 1.0.39, when running on a machine with Yarn 3.0 or above, Claude Code could have been tricked to execute code contained in a project via yarn plugins before the user accepted the startup trust dialog. Exploiting this would have required a user to start Claude Code in an untrusted directory and to be using Yarn 3.0 or above. This issue has been patched in version 1.0.39.

Claude Code is an agentic coding tool. Versions before 1.0.111 were vulnerable to Code Injection due to a bug in the startup trust dialog implementation. Claude Code could be tricked to execute code contained in a project before the user accepted the startup trust dialog. Exploiting this requires a user to start Claude Code in an untrusted directory. Users on standard Claude Code auto-update will have received this fix automatically. Users performing manual updates are advised to update to the latest version. This issue is fixed in version 1.0.111.

Claude Code is an agentic coding tool. At startup, Claude Code executed a command templated in with `git config user.email`. Prior to version 1.0.105, a maliciously configured user email in git could be used to trigger arbitrary code execution before a user accepted the workspace trust dialog. Users on standard Claude Code auto-update will have received this fix automatically. Users performing manual updates are advised to update to version 1.0.105 or the latest version.

Claude Code is an agentic coding tool. Prior to Claude Code version 1.0.39, when using Claude Code with Yarn versions 2.0+, Yarn plugins are auto-executed when running yarn --version. This could lead to a bypass of the directory trust dialog in Claude Code, as plugins would be executed prior to the user accepting the risks of working in an untrusted directory. Users running Yarn Classic were unaffected by this issue. This issue has been fixed in version 1.0.39. Users on standard Claude Code auto-update will have received this fix automatically. Users performing manual updates are advised to update to the latest version.

Claude Code is an agentic coding tool. Prior to 1.0.93, Due to errors in parsing shell commands related to $IFS and short CLI flags, it was possible to bypass the Claude Code read-only validation and trigger arbitrary code execution. Reliably exploiting this requires the ability to add untrusted content into a Claude Code context window. This vulnerability is fixed in 1.0.93.

Claude Code is an agentic coding tool. Prior to version 2.0.31, due to an error in sed command parsing, it was possible to bypass the Claude Code read-only validation and write to arbitrary files on the host system. This issue has been patched in version 2.0.31.

Claude Code is an agentic coding tool. In versions below 1.0.20, an error in command parsing makes it possible to bypass the Claude Code confirmation prompt to trigger execution of an untrusted command. Reliably exploiting this requires the ability to add untrusted content into a Claude Code context window. This is fixed in version 1.0.20.

Claude Code is an agentic coding tool. Prior to version 2.0.72, due to an error in command parsing, it was possible to bypass the Claude Code confirmation prompt to trigger execution of untrusted commands through the find command. Reliably exploiting this required the ability to add untrusted content into a Claude Code context window. This issue has been patched in version 2.0.72.

The com.phlox.tvwebbrowser TV Bro application through 2.0.0 for Android mishandles external intents through WebView. This allows attackers to execute arbitrary code, create arbitrary files. and perform arbitrary downloads via JavaScript that uses takeBlobDownloadData.

A vulnerability has been identified in Simcenter Amesim (All versions < V2021.1). The affected application contains a SOAP endpoint that could allow an unauthenticated remote attacker to perform DLL injection and execute arbitrary code in the context of the affected application process.

The MojoHaus Exec Maven plugin 1.1.1 for Maven allows code execution via a crafted XML document because a configuration element (within a plugin element) can specify an arbitrary program in an executable element (and can also specify arbitrary command-line arguments in an arguments element).

The npm module "shell-quote" 1.6.0 and earlier cannot correctly escape ">" and "<" operator used for redirection in shell. Applications that depend on shell-quote may also be vulnerable. A malicious user could perform code injection.

IBM HTTP Server 8.5, and 9.0 is vulnerable to denial of service and a potential remote code execution due to improper input validation.

The package convert-svg-core before 0.6.2 are vulnerable to Remote Code Injection via sending an SVG file containing the payload.

IBM WebSphere Application Server and IBM WebSphere Application Server Liberty - when using Intelligent Management with the WebSphere WebServer Plug-in component - are vulnerable to remote code execution and denial of service. This vulnerability can be exploited when an attacker impersonates backend servers and sends crafted responses to the plug-in.

Unsafe use of Python's eval() on server-received data in the vector_in() function in amazon-redshift-python-driver before 2.1.14 allows a rogue server or man-in-the-middle actor to execute arbitrary code on the client. To remediate this issue, users should upgrade to version 2.1.14.

IBM Web Server Plug-ins for WebSphere Application Server and WebSphere Liberty 8.5, 9.0 IBM WebSphere Application Server and WebSphere Application Server Liberty are vulnerable to remote code execution in the Web Server Plug-ins, through a specially crafted request.

A code injection vulnerability in the SeDebugPrivilege component of Trezor Bridge 2.0.27 allows attackers to escalate privileges.

In 0.10.0 or older versions of Apache Pinot, Pinot query endpoint and realtime ingestion layer has a vulnerability in unprotected environments due to a groovy function support. In order to avoid this, we disabled the groovy function support by default from Pinot release 0.11.0. See https://docs.pinot.apache.org/basics/releases/0.11.0

LLaMA-Factory through 0.9.5 contains a remote code execution vulnerability that allows attackers with WebUI access to execute arbitrary Python code by supplying a malicious model path in the Chat or Training interfaces. The application passes user-supplied model path input unvalidated into AutoTokenizer.from_pretrained() and AutoModel.from_pretrained() with a hardcoded trust_remote_code=True parameter, causing the Hugging Face transformers library to fetch and execute arbitrary code from a remote or local model repository with the privileges of the server process.

A flaw has been found in FoundationAgents MetaGPT up to 0.8.1. This vulnerability affects the function ActionNode.xml_fill of the file metagpt/actions/action_node.py of the component XML Handler. Executing a manipulation can lead to improper neutralization of directives in dynamically evaluated code. The attack may be launched remotely. The exploit has been published and may be used. The project was informed of the problem early through a pull request but has not reacted yet.

Tenda AC10 V1.0 V15.03.06.23 has a command injection vulnerablility located in the formexeCommand function. The str variable receives the cmdinput parameter from a POST request and is later assigned to the cmd_buf variable, which is directly used in the doSystemCmd function, causing an arbitrary command execution.

A vulnerability has been found in Fosowl agenticSeek 0.1.0. Impacted is the function PyInterpreter.execute of the file sources/tools/PyInterpreter.py of the component query Endpoint. Such manipulation leads to code injection. The attack can be launched remotely. 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 vulnerability was detected in FoundationAgents MetaGPT up to 0.8.1. This affects the function check_solution of the component HumanEvalBenchmark/MBPPBenchmark. Performing a manipulation results in code injection. The attack may be initiated remotely. The exploit is now public and may be used. The project was informed of the problem early through a pull request but has not reacted yet.

An arbitrary file upload vulnerability in the component /jquery-file-upload/server/php/index.php of Hospital Management System v4.0 allows an unauthenticated attacker to upload any file to the server and execute arbitrary code.

The events-manager plugin before 5.6 for WordPress has code injection.

Crawl4AI is an open-source LLM friendly web crawler & scraper. Prior to 0.8.7, the _safe_eval_expression() function in the computed fields feature uses an AST validator that only blocks attributes starting with underscore. Python generator and frame object attributes (gi_frame, f_back, f_builtins) do NOT start with underscore, enabling a complete sandbox escape to achieve arbitrary code execution. The attack requires no authentication (JWT disabled by default) and is triggered via POST /crawl with a crafted extraction schema. This vulnerability is fixed in 0.8.7.

An issue in mojoPortal v.2.7.0.0 allows a remote attacker to execute arbitrary code via a crafted script to the layout.master skin file at the Skin management component.

Codiad Web IDE through 2.8.4 allows PHP Code injection.

An issue in the loopback request handling component of fossar selfoss v2.20-SNAPSHOT allows attackers to execute arbitrary commands and obtain sensitive information via supplying a crafted HTTP request.

Eval injection in the Math plugin of Limnoria (before 2019.11.09) and Supybot (through 2018-05-09) allows remote unprivileged attackers to disclose information or possibly have unspecified other impact via the calc and icalc IRC commands.

An issue in the sendmail transport integration component of YouTransfer v1.0.6 allows attackers to execute arbitrary code via supplying a crafted request.

An OS command injection vulnerability in the media archiving and export pipeline component of kanishka-linux Reminiscence v0.3.0 allows attackers to execute arbitrary commands via supplying a crafted input.

YesWiki is a wiki system written in PHP. Prior to version 4.6.6, an unsafe execution vulnerability exists in the Bazar form field calculator (CalcField.php) of YesWiki. The application attempts to sanitize user-defined mathematical formulas using a complex recursive regular expression before passing them to the PHP eval() function. This implementation is inherently flawed: it is vulnerable to Regular Expression Denial of Service (ReDoS / Stack Overflow) which can crash the server, and it creates a high-risk architecture where any logic bypass directly results in arbitrary PHP code execution. Version 4.6.6 patches the issue.

baserCMS is a website development framework. In versions 4.6.0 through 4.7.6, there is a Code Injection vulnerability in the mail form of baserCMS. As of time of publication, no known patched versions are available.

Code injection in pluginconfig.php in Image Uploader and Browser for CKEditor before 4.1.9 allows remote authenticated users to execute arbitrary PHP code.

The Angular Language Service VS Code Extension provides a rich editing experience for Angular templates. Prior to 21.2.4, the client-side Angular Language Service VS Code extension reads the custom TypeScript SDK paths typescript.tsdk and js/ts.tsdk.path directly from workspace configurations (.vscode/settings.json) without verifying VS Code Workspace Trust state or asking for user consent (located in client/src/client.ts). The client-side extension then passes the parsed settings path as a command-line argument (--tsdk) to the background Node.js language server process. During server initialization, the background language server resolves and dynamically imports (via standard Node.js require()) the module library tsserverlibrary.js relative to the workspace-specified custom directory path. An attacker can exploit this behavior by committing a repository containing a local malicious tsserverlibrary.js script inside a custom folder, and a crafted .vscode/settings.json file pointing to that folder. When a developer opens the repository folder in VS Code, the extension automatically attempts to initialize and load the server, which dynamically resolves, loads, and executes the malicious script silently in the background. This vulnerability is fixed in 21.2.4.

An issue in mystenlabs Sui Blockchain before v.1.6.3 allow a remote attacker to execute arbitrary code and cause a denial of service via a crafted compressed script to the Sui node component.

The Imou Life com.mm.android.smartlifeiot application through 6.8.0 for Android allows Remote Code Execution via a crafted intent to an exported component. This relates to the com.mm.android.easy4ip.MainActivity activity. JavaScript execution is enabled in the WebView, and direct web content loading occurs.

dst-admin v1.5.0 was discovered to contain a remote command execution (RCE) vulnerability via the userId parameter at /home/playerOperate.

Versions of the package simple-git before 3.16.0 are vulnerable to Remote Code Execution (RCE) via the clone(), pull(), push() and listRemote() methods, due to improper input sanitization. This vulnerability exists due to an incomplete fix of [CVE-2022-25912](https://security.snyk.io/vuln/SNYK-JS-SIMPLEGIT-3112221).

qibosoft 7 allows remote code execution because do/jf.php makes eval calls. The attacker can use the Point Introduction Management feature to supply PHP code to be evaluated. Alternatively, the attacker can access admin/index.php?lfj=jfadmin&action=addjf via CSRF, as demonstrated by a payload in the content parameter.

OpenMed before 1.5.2 contains a remote code execution vulnerability in the PII privacy-filter model loading path. The privacy-filter dispatcher used broad substring matching on the user-supplied model_name parameter, allowing a value such as attacker/foo-privacy-filter-bar to route through a path that loads Hugging Face models with trust_remote_code=True. An unauthenticated attacker can supply a malicious model repository containing custom Transformers code via auto_map in config.json or tokenizer_config.json, which is imported and executed with the privileges of the OpenMed service process.

main.py in Searchor before 2.4.2 uses eval on CLI input, which may cause unexpected code execution.

A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.1. This affects the function generate_thoughts of the file metagpt/strategy/tot.py of the component Tree-of-Thought Solver. The manipulation leads to code injection. It is possible to initiate the attack remotely. The exploit is publicly available and might be used. The project was informed of the problem early through an issue report but has not responded yet.

A vulnerability, which was classified as problematic, was found in wix-incubator jam up to e87a6fd85cf8fb5ff37b62b2d68f917219d07ae9. This affects an unknown part of the file jam.py of the component Jinja2 Template Handler. The manipulation of the argument config['template'] leads to improper neutralization of special elements used in a template engine. It is possible to launch the attack on the local host. The exploit has been disclosed to the public and may be used. This product takes the approach of rolling releases to provide continious delivery. Therefore, version details for affected and updated releases are not available.

parserIfLabel in inc/zzz_template.php in ZZZCMS zzzphp 1.7.3 allows remote attackers to execute arbitrary code because the danger_key function can be bypassed via manipulations such as strtr.

OneVision Workspace before WS23.1 SR1 (build w31.040) allows arbitrary Java EL execution.

Pi.Alert is a WIFI / LAN intruder detector with web service monitoring. Prior to 2026-05-07, Pi.Alert's SaveConfigFile() endpoint writes user-supplied numeric config values (e.g., SMTP_PORT) directly into pialert.conf without validation. Since pialert.conf is loaded via Python's exec() every 3–5 minutes by the background cron process, an attacker can inject arbitrary Python code and achieve unauthenticated OS-level RCE. On default installations (PIALERT_WEB_PROTECTION = False), no credentials are required. This vulnerability is fixed in 2026-05-07.

MCP Calculate Server is a mathematical calculation service based on MCP protocol and SymPy library. Prior to 0.1.1, the use of eval() to evaluate mathematical expressions without proper input sanitization leads to remote code execution. This vulnerability is fixed in 0.1.1.