Weakness (CWE)
CWE-95: Improper Neutralization of Directives in Dynamically Evaluated Code
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").
How to prevent it (4 mitigations)
-
Architecture and Design, Implementation
If possible, refactor your code so that it does not need to use eval() at all.
-
Implementation
Assume all input is malicious. Use an "accept known good" input validation strategy, i.e., use a list of acceptable inputs that strictly conform to specifications. Reject any input that does not strictly conform to specifications, or transform it into something that does.
When performing input validation, consider all potentially relevant properties, including length, type of input, the full range of acceptable values, missing or extra inputs, syntax, consistency across related fields, and conformance to business rules. As an example of business rule logic, "boat" may be syntactically valid because it only contains alphanumeric characters, but it is not valid if the input is only expected to contain colors such as "red" or "blue."
Do not rely exclusively on looking for malicious or malformed inputs. This is likely to miss at least one undesirable input, especially if the code's environment changes. This can give attackers enough room to bypass the intended validation. However, denylists can be useful for detecting potential attacks or determining which inputs are so malformed that they should be rejected outright.
-
Implementation
Inputs should be decoded and canonicalized to the application's current internal representation before being validated (CWE-180, CWE-181). Make sure that your application does not inadvertently decode the same input twice (CWE-174). Such errors could be used to bypass allowlist schemes by introducing dangerous inputs after they have been checked. Use libraries such as the OWASP ESAPI Canonicalization control.
Consider performing repeated canonicalization until your input does not change any more. This will avoid double-decoding and similar scenarios, but it might inadvertently modify inputs that are allowed to contain properly-encoded dangerous content.
-
Implementation
For Python programs, it is frequently encouraged to use the ast.literal_eval() function instead of eval, since it is intentionally designed to avoid executing code. However, an adversary could still cause excessive memory or stack consumption via deeply nested structures [REF-1372], so the python documentation discourages use of ast.literal_eval() on untrusted data [REF-1373].
Source: MITRE CWE-95
32 results · Clear filters · Feed
-
Medium 6.6
Chamilo LMS: Conditional PHP execution through vendored QuickForm eval
Chamilo LMS is an open-source learning management system. In the 1.x line before 1.11.40, and the 2.x line before 2.0.1, multiple `eval()` calls in the vendored PEAR `HTML_QuickForm` libraries can execute arbitrary PHP code if form element names contain attac…
-
Medium 4.3
Crawl4AI 0.9.0 before 0.9.5 JavaScript Injection via css_selector and wait_for
Crawl4AI from 0.9.0 before 0.9.5 contains a JavaScript injection vulnerability in the Docker API server that allows authenticated API clients to run arbitrary JavaScript in the crawl page via allowlisted css_selector and wait_for fields. Attackers can inject …
-
Medium 6.6
Server-Side Script Injection in Manager Web UI in Progress WS_FTP Server
Improper neutralization of directives in dynamically evaluated code in the Manager web console of Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows could allow an authenticated Manager administrator to execute arbitrary comman…
-
Medium 5.3
THU-MIG yoloe AutoBackend Model Loading autobackend.py eval eval injection
A security flaw has been discovered in THU-MIG yoloe up to 40cd606cabdbe2b566d6f14a6b162c89206e9a1b. This affects the function eval of the file ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. The manipulation of the argument imgsz/na…
-
Medium 5.3
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.get_modelmeta.custom_metadata_map of the file ultralytics/nn/autobackend.py of the component AutoBackend Model Loading. Thi…
-
Medium 4.8
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.get_modelmeta.custom_metadata_map of the file models/MobileSAMv2/ultralytics/nn/autobackend.py of the component AutoBackend…
-
Medium 4.7
django-haystack more_like_this Template Tag elasticsearch_backend.py _to_python eval injection
A vulnerability has been found in django-haystack up to 3.3.0. Affected is the function _to_python of the file haystack/backends/elasticsearch_backend.py of the component more_like_this Template Tag Handler. Such manipulation of the argument result_class lead…
-
Medium 4.1
Langflow Code Execution via eval() in Component Input Schema
Langflow 1.0.16 before 1.12.0 and 0.0.94 before 1.12.0 contain an unsafe eval() vulnerability in schema.py that allows authenticated attackers to achieve code execution by placing a Python object with a malicious __repr__ method into component input options l…
-
Medium 5.3
Information Disclosure vulnerability in SAP Fiori Launchpad
SAP Fiori Launchpad does not sufficiently validate certain user-controlled input. An unauthenticated attacker could craft a malicious link that, when clicked by an authenticated user, causes the browser to load attacker-controlled content from an external loc…
-
Medium 6.6
Improper Input Handling in Okta Access Gateway Management Console Exception Handler
The Okta Access Gateway management console passes user-supplied input to eval() without sanitization during an authenticated administrator SSH session. As a result, the unsanitized input is executed directly, leading to code execution with the privileges of t…
-
Medium 5.1
Vim: Vimscript Code Injection in cucumber filetype plugin via crafted step-definition regex
Vim is an open source, command line text editor. Prior to version 9.2.0496, a code injection vulnerability exists in s:stepmatch() in the cucumber filetype plugin (runtime/ftplugin/cucumber.vim) on Vim builds with +ruby support. Step-definition patterns read …
-
Medium 5.6
ByteDance verl grader.py math_equal sandbox
A vulnerability was identified in ByteDance verl up to 0.7.0. Affected is the function math_equal of the file prime_math/grader.py. The manipulation leads to sandbox issue. It is possible to initiate the attack remotely. The complexity of an attack is rather …
-
Medium 4.7
Pagekit CMS StringStorage Template PhpEngine.php evaluate eval injection
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…
-
Medium 6.9
Stored XSS via Eval Injection in EchartsRander Component
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 th…
-
Medium 6.6
Eval Injection in Rapid7 Insight Agent
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 c…
-
Medium 5.9
LAN Code Execution on TP-Link Archer MR200, Archer C20, TL-WR850N and TL-WR845N
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-Mi…
-
Medium 6.1
PHP-Fusion 9.03.50 - 'panels.php' Eval Injection
PHP-Fusion 9.03.50 contains a remote code execution vulnerability in the 'add_panel_form()' function that allows attackers to execute arbitrary code through an eval() function with unsanitized POST data. Attackers can exploit the vulnerability by sending craf…
-
Medium 5.4
Command Injection in nvm via NVM_AUTH_HEADER in wget code path
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 w…
-
Medium 5.3
Dioxus Components has JavaScript injection via user-supplied IDs
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 41e4242ecb1062d04ae42a…
-
Medium 6.4CVE-2026-23885 Public exploit
AlchemyCMS has Authenticated Remote Code Execution (RCE) via eval injection in ResourcesHelper
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 i…
-
Medium 5.5
An injection issue was addressed with improved validation. This issue is fixed in macOS Tahoe 26.1. An app may be able …
-
Medium 6.5
TOTOLINK A3002R v4.0.0-B20230531.1404 was discovered to contain an eval injection vulnerability via the eval() function.
-
Medium 5.5
letta-ai letta interface.py function_message eval injection
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 …
-
Medium 4.4
conda-forge-ci-setup Allows Arbitrary Code Execution via Insecure Version Parsing
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.y…
-
Medium 6.3
OZI-Project/ozi-publish Code Injection vulnerability
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 c…
-
Medium —
Butterfly's parseJSON, getJSON functions eval malicious input, leading to remote code execution (RCE)
### Summary Usage of the `Butterfly.prototype.parseJSON` or `getJSON` functions on an attacker-controlled crafted input string allows the attacker to execute arbitrary JavaScript code on the server. Since Butterfly JavaScript code has access to Java classes, …
-
Medium 5.3
vyper performs double eval of the argument of sqrt
Vyper is a pythonic Smart Contract Language for the Ethereum virtual machine. In versions 0.3.10 and prior, using the `sqrt` builtin can result in double eval vulnerability when the argument has side-effects. It can be seen that the `build_IR` function of the…
-
Medium 5.3
vyper performs double eval of raw_args in create_from_blueprint
Vyper is a pythonic Smart Contract Language for the Ethereum virtual machine. In versions 0.3.10 and prior, using the `create_from_blueprint` builtin can result in a double eval vulnerability when `raw_args=True` and the `args` argument has side-effects. It c…
-
Medium 4.9
Authenticated eval injection in B. Braun Space Battery pack SP with Wi-Fi
An improper neutralization of directives in dynamically evaluated code vulnerability in the WiFi Battery embedded web server in versions L90/U70 and L92/U92 can be used to gain administrative access to the WiFi communication module. An authenticated user, hav…
-
Medium 6.1
Code injection issue in Portal for ArcGIS (10.7.1 and 10.8.1)
There is a code injection vulnerability in Esri Portal for ArcGIS versions 10.8.1 and below that may allow a remote, unauthenticated attacker to pass strings which could potentially cause arbitrary code execution.
-
Medium 6.5
A function module of SAP NetWeaver AS ABAP (Reconciliation Framework), versions - 700, 701, 702, 710, 711, 730, 731, 74…
-
Medium 4.4
Directive injection when using dynamic overrides with user input in RubyGems secure_headers
In Secure Headers (RubyGem secure_headers), a directive injection vulnerability is present in versions before 3.8.0, 5.1.0, and 6.2.0. If user-supplied input was passed into append/override_content_security_policy_directives, a semicolon could be injected lea…