CWE-95
Improper Neutralization of Directives in Dynamically Evaluated Code ('Eval Injection')
VariantIncompleteLikelihood: Medium
Description
The product receives input from an upstream component, but it does not neutralize or incorrectly neutralizes code syntax before using the input in a dynamic evaluation call (e.g. "eval").
Hierarchy (View 1000)
Parents
Children
none
Related attack patterns (CAPEC)
CAPEC-35
CVEs mapped to this weakness (37)
page 2 of 2| CVE | Sev | Risk | CVSS | EPSS | KEV | Published | Description |
|---|---|---|---|---|---|---|---|
| CVE-2025-58365 | Hig | 0.50 | — | 0.01 | Sep 8, 2025 | The XWiki blog application allows users of the XWiki platform to create and manage blog posts. Prior to version 9.14, the blog application in XWiki allowed remote code execution for any user who has edit right on any page. Normally, these are all logged-in users as they can edit their own user profile. For an exploit, it is sufficient to add an object of type `Blog.BlogPostClass` to any page and to add some script macro with the exploit code to the "Content" field of that object. The vulnerability has been patched in the blog application version 9.14 by executing the content of blog posts with the rights of the appropriate author. No known workarounds are available. | |
| CVE-2026-42079 | Hig | 0.49 | 8.6 | 0.00 | May 4, 2026 | PPTAgent is an agentic framework for reflective PowerPoint generation. Prior to commit 418491a, PPTAgent is vulnerable to arbitrary code execution via Python eval() of LLM-generated code with builtins in scope. This issue has been patched via commit 418491a. | |
| CVE-2026-5971 | Hig | 0.47 | 7.3 | 0.00 | Apr 9, 2026 | 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. | |
| CVE-2026-22666 | Hig | 0.47 | 7.2 | 0.00 | Apr 7, 2026 | Dolibarr ERP/CRM versions prior to 23.0.2 contain an authenticated remote code execution vulnerability in the dol_eval_standard() function that fails to apply forbidden string checks in whitelist mode and does not detect PHP dynamic callable syntax. Attackers with administrator privileges can inject malicious payloads through computed extrafields or other evaluation paths using PHP dynamic callable syntax to bypass validation and achieve arbitrary command execution via eval(). | |
| CVE-2026-4965 | Hig | 0.47 | 7.3 | 0.00 | Mar 27, 2026 | A vulnerability was detected in letta-ai letta 0.16.4. This issue affects the function resolve_type of the file letta/functions/ast_parsers.py of the component Incomplete Fix CVE-2025-6101. Performing a manipulation results in improper neutralization of directives in dynamically evaluated code. The attack can be initiated remotely. The exploit is now public and may be used. The vendor was contacted early about this disclosure but did not respond in any way. | |
| CVE-2024-10633 | Hig | 0.47 | 7.3 | 0.00 | Jan 26, 2025 | The Quiz Maker Business, Developer, and Agency plugins for WordPress is vulnerable to arbitrary shortcode execution in all versions up to, and including, 8.8.0 (Business), up to, and including, 21.8.0 (Developer), and up to, and including, 31.8.0 (Agency). This is due to the software allowing users to execute an action that does not properly validate a value before running do_shortcode. This makes it possible for unauthenticated attackers to execute arbitrary shortcodes. | |
| CVE-2024-45858 | Hig | 0.44 | 7.8 | 0.00 | Sep 18, 2024 | An arbitrary code execution vulnerability exists in versions 0.2.9 up to 0.5.10 of the Guardrails AI Guardrails framework because of the way it validates XML files. If a victim user loads a maliciously crafted XML file containing Python code, the code will be passed to an eval function, causing it to execute on the user's machine. | |
| CVE-2026-4837 | Med | 0.43 | 6.6 | 0.00 | Apr 8, 2026 | An eval() injection vulnerability in the Rapid7 Insight Agent beaconing logic for Linux versions could theoretically allow an attacker to achieve remote code execution as root via a crafted beacon response. Because the Agent uses mutual TLS (mTLS) to verify commands from the Rapid7 Platform, it is unlikely that the eval() function could be exploited remotely without prior, highly privileged access to the backend platform. | |
| CVE-2026-23885 | Med | 0.42 | 6.4 | 0.00 | Jan 19, 2026 | Alchemy is an open source content management system engine written in Ruby on Rails. Prior to versions 7.4.12 and 8.0.3, the application uses the Ruby `eval()` function to dynamically execute a string provided by the `resource_handler.engine_name` attribute in `Alchemy::ResourcesHelper#resource_url_proxy`. The vulnerability exists in `app/helpers/alchemy/resources_helper.rb` at line 28. The code explicitly bypasses security linting with `# rubocop:disable Security/Eval`, indicating that the use of a dangerous function was known but not properly mitigated. Since `engine_name` is sourced from module definitions that can be influenced by administrative configurations, it allows an authenticated attacker to escape the Ruby sandbox and execute arbitrary system commands on the host OS. Versions 7.4.12 and 8.0.3 fix the issue by replacing `eval()` with `send()`. | |
| CVE-2025-15551 | Med | 0.36 | 5.6 | 0.00 | Feb 5, 2026 | The response coming from TP-Link Archer MR200 v5.2, C20 v5 and v6, TL-WR850N v3, and TL-WR845N v4 for any request is getting executed by the JavaScript function like eval directly without any check. Attackers can exploit this vulnerability via a Man-in-the-Middle (MitM) attack to execute JavaScript code on the router's admin web portal without the user's permission or knowledge. | |
| CVE-2025-6101 | Med | 0.36 | 5.5 | 0.00 | Jun 16, 2025 | A vulnerability classified as critical has been found in letta-ai letta up to 0.4.1. Affected is the function function_message of the file letta/letta/interface.py. The manipulation of the argument function_name/function_args leads to improper neutralization of directives in dynamically evaluated code. The exploit has been disclosed to the public and may be used. | |
| CVE-2026-1665 | Med | 0.35 | — | 0.00 | Jan 29, 2026 | A command injection vulnerability exists in nvm (Node Version Manager) versions 0.40.3 and below. The nvm_download() function uses eval to execute wget commands, and the NVM_AUTH_HEADER environment variable was not sanitized in the wget code path (though it was sanitized in the curl code path). An attacker who can set environment variables in a victim's shell environment (e.g., via malicious CI/CD configurations, compromised dotfiles, or Docker images) can inject arbitrary shell commands that execute when the victim runs nvm commands that trigger downloads, such as 'nvm install' or 'nvm ls-remote'. | |
| CVE-2026-24474 | Med | 0.34 | — | 0.00 | Jan 24, 2026 | Dioxus Components is a shadcn-style component library for the Dioxus app framework. Prior to commit 41e4242ecb1062d04ae42a5215363c1d9fd4e23a, `use_animated_open` formats a string for `eval` with an `id` that can be user supplied. Commit 41e4242ecb1062d04ae42a5215363c1d9fd4e23a patches the issue. | |
| CVE-2025-47271 | Med | 0.34 | — | 0.00 | May 12, 2025 | The OZI action is a GitHub Action that publishes releases to PyPI and mirror releases, signature bundles, and provenance in a tagged release. In versions 1.13.2 through 1.13.5, potentially untrusted data flows into PR creation logic. A malicious actor could construct a branch name that injects arbitrary code. This is patched in 1.13.6. As a workaround, one may downgrade to a version prior to 1.13.2. | |
| CVE-2026-6652 | Med | 0.31 | 4.7 | 0.00 | Apr 20, 2026 | A weakness has been identified in Pagekit CMS up to 1.0.18. This issue affects the function evaluate of the file app/modules/view/src/PhpEngine.php of the component StringStorage Template Handler. This manipulation causes improper neutralization of directives in dynamically evaluated code. Remote exploitation of the attack is possible. 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. | |
| CVE-2026-39423 | Med | 0.28 | 5.4 | 0.00 | Apr 14, 2026 | MaxKB is an open-source AI assistant for enterprise. Versions 2.7.1 and below contain an Eval Injection vulnerability in the Markdown rendering engine that allows any user capable of interacting with the AI chat interface to execute arbitrary JavaScript in the browsers of other users, including administrators, resulting in Stored Cross-Site Scripting (XSS). This issue has been fixed in version 2.8.0. | |
| CVE-2025-49598 | Med | 0.22 | — | 0.00 | Jun 13, 2025 | conda-forge-ci-setup is a package installed by conda-forge each time a build is run on CI. The conda-forge-ci-setup-feedstock setup script is vulnerable due to the unsafe use of the eval function when parsing version information from a custom-formatted meta.yaml file. An attacker controlling meta.yaml can inject malicious code into the version assignment, which is executed during file processing, leading to arbitrary code execution. Exploitation requires an attacker to modify the recipe file by manipulating the RECIPE_DIR variable and introducing a malicious meta.yaml file. While this is more feasible in CI/CD pipelines, it is uncommon in typical environments, reducing overall risk. This vulnerability is fixed in 4.15.0. |