CWE-94
Allowed-with-ReviewImproper Control of Generation of Code ('Code Injection')
Abstraction: Base · Status: Draft
The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.
9496 vulnerabilities reference this CWE, most recent first.
GHSA-7FW9-72Q7-3G7J
Vulnerability from github – Published: 2026-08-27 21:31 – Updated: 2026-09-01 21:31ServiceNow has remediated a sandbox escape security issue that was identified in the Now Platform. This security issue could allow an unauthenticated user to execute arbitrary code within the Now Platform, potentially leading to more access to the Now Platform than intended.
ServiceNow deployed a security update to hosted instances and ServiceNow provided the update to our partners and self-hosted customers. We are not currently aware of malicious exploitation against ServiceNow instances.
We recommend customers promptly apply appropriate updates or upgrade to a patched release if they have not already done so.
{
"affected": [],
"aliases": [
"CVE-2026-6876"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-08-27T20:18:32Z",
"severity": "HIGH"
},
"details": "ServiceNow has remediated a sandbox escape security issue that was identified in the Now Platform. This security issue could allow an unauthenticated user to execute arbitrary code within the Now Platform, potentially leading to more access to the Now Platform than intended.\u00a0\u00a0\n\n\n\n\n\nServiceNow deployed a security update to hosted instances and ServiceNow provided the update to our partners and self-hosted customers. We are not currently aware of malicious exploitation against ServiceNow instances.\u00a0\n\n\n\nWe recommend customers promptly apply appropriate updates or upgrade to a patched release if they have not already done so.",
"id": "GHSA-7fw9-72q7-3g7j",
"modified": "2026-09-01T21:31:27Z",
"published": "2026-08-27T21:31:52Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-6876"
},
{
"type": "WEB",
"url": "https://support.servicenow.com/kb?id=kb_article_view\u0026sysparm_article=KB3152242"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
"type": "CVSS_V4"
}
]
}
GHSA-7G4G-9H43-3XVC
Vulnerability from github – Published: 2022-05-17 01:28 – Updated: 2022-05-17 01:28The System Status Collection Daemon (SSCD) in Cisco TelePresence System 500-37, 1000, 1300-65, and 3xxx before 1.10.2(42), and 500-32, 1300-47, TX1310 65, and TX9xxx before 6.0.4(11), allows remote attackers to execute arbitrary commands or cause a denial of service (stack memory corruption) via a crafted XML-RPC message, aka Bug ID CSCui32796.
{
"affected": [],
"aliases": [
"CVE-2014-0661"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2014-01-22T21:55:00Z",
"severity": "HIGH"
},
"details": "The System Status Collection Daemon (SSCD) in Cisco TelePresence System 500-37, 1000, 1300-65, and 3xxx before 1.10.2(42), and 500-32, 1300-47, TX1310 65, and TX9xxx before 6.0.4(11), allows remote attackers to execute arbitrary commands or cause a denial of service (stack memory corruption) via a crafted XML-RPC message, aka Bug ID CSCui32796.",
"id": "GHSA-7g4g-9h43-3xvc",
"modified": "2022-05-17T01:28:45Z",
"published": "2022-05-17T01:28:45Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2014-0661"
},
{
"type": "WEB",
"url": "https://exchange.xforce.ibmcloud.com/vulnerabilities/90624"
},
{
"type": "WEB",
"url": "http://osvdb.org/102362"
},
{
"type": "WEB",
"url": "http://secunia.com/advisories/56533"
},
{
"type": "WEB",
"url": "http://tools.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-20140122-cts"
},
{
"type": "WEB",
"url": "http://www.securityfocus.com/bid/65071"
},
{
"type": "WEB",
"url": "http://www.securitytracker.com/id/1029656"
}
],
"schema_version": "1.4.0",
"severity": []
}
GHSA-7G4V-8CVX-GP5Q
Vulnerability from github – Published: 2024-07-05 15:32 – Updated: 2025-02-13 18:32The CloudStack cluster service runs on unauthenticated port (default 9090) that can be misused to run arbitrary commands on targeted hypervisors and CloudStack management server hosts. Some of these commands were found to have command injection vulnerabilities that can result in arbitrary code execution via agents on the hosts that may run as a privileged user. An attacker that can reach the cluster service on the unauthenticated port (default 9090), can exploit this to perform remote code execution on CloudStack managed hosts and result in complete compromise of the confidentiality, integrity, and availability of CloudStack managed infrastructure.
Users are recommended to restrict the network access to the cluster service port (default 9090) on a CloudStack management server host to only its peer CloudStack management server hosts. Users are recommended to upgrade to version 4.18.2.1, 4.19.0.2 or later, which addresses this issue.
{
"affected": [],
"aliases": [
"CVE-2024-38346"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-07-05T14:15:02Z",
"severity": "CRITICAL"
},
"details": "The CloudStack cluster service runs on unauthenticated port (default 9090) that can be misused to run arbitrary commands on targeted hypervisors and CloudStack management server hosts. Some of these commands were found to have command injection vulnerabilities that can result in arbitrary code execution via agents on the hosts that may run as a privileged user.\u00a0An attacker that can reach the cluster service on the unauthenticated\u00a0port (default 9090), can exploit this to perform remote code execution on CloudStack managed hosts and result in complete\u00a0compromise of the confidentiality, integrity, and availability of CloudStack managed infrastructure.\n\nUsers are recommended to restrict the network access to the cluster service port (default 9090) on a CloudStack management server host to only its peer CloudStack management server hosts.\u00a0Users are recommended to upgrade to version 4.18.2.1, 4.19.0.2 or later, which addresses this issue.\n\n",
"id": "GHSA-7g4v-8cvx-gp5q",
"modified": "2025-02-13T18:32:31Z",
"published": "2024-07-05T15:32:06Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2024-38346"
},
{
"type": "WEB",
"url": "https://cloudstack.apache.org/blog/security-release-advisory-4.19.0.2-4.18.2.1"
},
{
"type": "WEB",
"url": "https://lists.apache.org/thread/6l51r00csrct61plkyd3qg3fj99215d1"
},
{
"type": "WEB",
"url": "https://www.shapeblue.com/shapeblue-security-advisory-apache-cloudstack-security-releases-4-18-2-1-and-4-19-0-2"
},
{
"type": "WEB",
"url": "http://www.openwall.com/lists/oss-security/2024/07/05/1"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-7G53-JJ25-JHGR
Vulnerability from github – Published: 2023-03-06 21:30 – Updated: 2023-03-14 17:59Funadmin v3.2.0 was discovered to contain a remote code execution (RCE) vulnerability via the component \controller\Addon.php.
{
"affected": [
{
"package": {
"ecosystem": "Packagist",
"name": "funadmin/funadmin"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "3.2.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2023-24776"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": true,
"github_reviewed_at": "2023-03-14T17:59:01Z",
"nvd_published_at": "2023-03-06T20:15:00Z",
"severity": "CRITICAL"
},
"details": "Funadmin v3.2.0 was discovered to contain a remote code execution (RCE) vulnerability via the component \\controller\\Addon.php.",
"id": "GHSA-7g53-jj25-jhgr",
"modified": "2023-03-14T17:59:01Z",
"published": "2023-03-06T21:30:18Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2023-24776"
},
{
"type": "WEB",
"url": "https://github.com/funadmin/funadmin/issues/7"
},
{
"type": "PACKAGE",
"url": "https://github.com/funadmin/funadmin"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
],
"summary": "Remote code execution in Funadmin"
}
GHSA-7G5W-PQ96-8C5W
Vulnerability from github – Published: 2026-05-11 18:31 – Updated: 2026-05-18 15:22The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 (2025-13-04) contains an insecure deserialization vulnerability (CWE-502) in its checkpoint loading mechanism. The load_checkpoint() function in checkpoint.py and the checkpoint loading code in eval.py use torch.load() without enabling the security-restrictive weights_only=True parameter. This allows the deserialization of arbitrary Python objects via the pickle module. An attacker can exploit this by providing a maliciously crafted checkpoint file. When a victim loads this checkpoint during model warmstarting or evaluation, arbitrary code is executed on the victim's system.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "flash_attn"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "2.8.3"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-31253"
],
"database_specific": {
"cwe_ids": [
"CWE-502",
"CWE-94"
],
"github_reviewed": true,
"github_reviewed_at": "2026-05-18T15:22:20Z",
"nvd_published_at": "2026-05-11T17:16:20Z",
"severity": "HIGH"
},
"details": "The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 (2025-13-04) contains an insecure deserialization vulnerability (CWE-502) in its checkpoint loading mechanism. The load_checkpoint() function in checkpoint.py and the checkpoint loading code in eval.py use torch.load() without enabling the security-restrictive weights_only=True parameter. This allows the deserialization of arbitrary Python objects via the pickle module. An attacker can exploit this by providing a maliciously crafted checkpoint file. When a victim loads this checkpoint during model warmstarting or evaluation, arbitrary code is executed on the victim\u0027s system.",
"id": "GHSA-7g5w-pq96-8c5w",
"modified": "2026-05-18T15:22:20Z",
"published": "2026-05-11T18:31:44Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-31253"
},
{
"type": "PACKAGE",
"url": "https://github.com/Dao-AILab/flash-attention"
},
{
"type": "WEB",
"url": "https://www.notion.so/CVE-2026-31253-35d1e1393188813f9e77e2038104bc49"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L",
"type": "CVSS_V3"
}
],
"summary": "flash-attention contains an insecure deserialization vulnerability in its checkpoint loading mechanism"
}
GHSA-7G8F-4JRF-X782
Vulnerability from github – Published: 2026-09-06 12:30 – Updated: 2026-09-06 12:30Bifrost HTTP transport before 2.0.0 accepts an enabled custom plugin whose path is an HTTP URL through unauthenticated POST /api/plugins when management authentication is disabled (the default, governance.auth_config.is_enabled=false). The shared-object loader treats an http-prefixed path as a download URL, writes the body to a temporary .so, and passes it to Go's plugin.Open. After a successful open, optional Init runs immediately with the supplied config as the Bifrost process user. On documented dynamically linked builds (DYNAMIC=1 / no static-link flags), which the vendor requires for custom Go plugins, plugin.Open is expected to succeed and this is unauthenticated remote code execution. On the published statically linked Docker image, plugin.Open fails with Dynamic loading not supported, so that build class is only server-side request forgery. Attack complexity is High because the attacker cannot force RCE on the default static image and a loadable plugin must match the host Go version, OS, architecture, and linkage. The 1.6.x HTTP transport line through 1.6.11 does not contain the fix.
{
"affected": [],
"aliases": [
"CVE-2026-86242"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-06T12:17:15Z",
"severity": "HIGH"
},
"details": "Bifrost HTTP transport before 2.0.0 accepts an enabled custom plugin whose path is an HTTP URL through unauthenticated POST /api/plugins when management authentication is disabled (the default, governance.auth_config.is_enabled=false). The shared-object loader treats an http-prefixed path as a download URL, writes the body to a temporary .so, and passes it to Go\u0027s plugin.Open. After a successful open, optional Init runs immediately with the supplied config as the Bifrost process user. On documented dynamically linked builds (DYNAMIC=1 / no static-link flags), which the vendor requires for custom Go plugins, plugin.Open is expected to succeed and this is unauthenticated remote code execution. On the published statically linked Docker image, plugin.Open fails with Dynamic loading not supported, so that build class is only server-side request forgery. Attack complexity is High because the attacker cannot force RCE on the default static image and a loadable plugin must match the host Go version, OS, architecture, and linkage. The 1.6.x HTTP transport line through 1.6.11 does not contain the fix.",
"id": "GHSA-7g8f-4jrf-x782",
"modified": "2026-09-06T12:30:23Z",
"published": "2026-09-06T12:30:23Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/maximhq/bifrost/security/advisories/GHSA-2qp8-4xgm-fw6g"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-86242"
},
{
"type": "WEB",
"url": "https://github.com/maximhq/bifrost/pull/5763"
},
{
"type": "WEB",
"url": "https://github.com/maximhq/bifrost/commit/e0057ff355f831c251eabe9d0e44f3a3748532c6"
},
{
"type": "WEB",
"url": "https://github.com/maximhq/bifrost"
},
{
"type": "WEB",
"url": "https://github.com/maximhq/bifrost/releases/tag/transports/v2.0.0"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
]
}
GHSA-7G93-8R7H-W6JJ
Vulnerability from github – Published: 2026-09-16 21:32 – Updated: 2026-09-16 21:32Under certain circumstances on affected platforms running Arista EOS with gRPC Network Packet Sampling Interface (gNPSI) enabled, an unauthenticated gNPSI client can craft a malicious request to allow arbitrary code execution, granting an attacker full administrative control over the compromised switch.
{
"affected": [],
"aliases": [
"CVE-2026-73456"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-16T19:17:32Z",
"severity": "CRITICAL"
},
"details": "Under certain circumstances on affected platforms running Arista EOS with gRPC Network Packet Sampling Interface (gNPSI) enabled, an unauthenticated gNPSI client can craft a malicious request to allow arbitrary code execution, granting an attacker full administrative control over the compromised switch.",
"id": "GHSA-7g93-8r7h-w6jj",
"modified": "2026-09-16T21:32:43Z",
"published": "2026-09-16T21:32:43Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-73456"
},
{
"type": "WEB",
"url": "https://www.arista.com/en/support/advisories-notices/security-advisory/24714-security-advisory-0158"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
"type": "CVSS_V4"
}
]
}
GHSA-7GCP-2GMQ-W3XH
Vulnerability from github – Published: 2022-05-13 01:38 – Updated: 2023-03-09 00:37RubyGems prior to 2.6.13 is vulnerable to maliciously crafted gem specifications that include terminal escape characters. Printing the gem specification would execute terminal escape sequences.
{
"affected": [
{
"package": {
"ecosystem": "RubyGems",
"name": "rubygems-update"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2.6.13"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2017-0899"
],
"database_specific": {
"cwe_ids": [
"CWE-150",
"CWE-94"
],
"github_reviewed": true,
"github_reviewed_at": "2023-03-09T00:37:49Z",
"nvd_published_at": "2017-08-31T20:29:00Z",
"severity": "CRITICAL"
},
"details": "RubyGems prior to 2.6.13 is vulnerable to maliciously crafted gem specifications that include terminal escape characters. Printing the gem specification would execute terminal escape sequences.",
"id": "GHSA-7gcp-2gmq-w3xh",
"modified": "2023-03-09T00:37:49Z",
"published": "2022-05-13T01:38:25Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2017-0899"
},
{
"type": "WEB",
"url": "https://github.com/rubygems/rubygems/commit/1bcbc7fe637b03145401ec9c094066285934a7f1"
},
{
"type": "WEB",
"url": "https://github.com/rubygems/rubygems/commit/ef0aa611effb5f54d40c7fba6e8235eb43c5a491"
},
{
"type": "WEB",
"url": "https://hackerone.com/reports/226335"
},
{
"type": "WEB",
"url": "https://access.redhat.com/errata/RHSA-2017:3485"
},
{
"type": "WEB",
"url": "https://access.redhat.com/errata/RHSA-2018:0378"
},
{
"type": "WEB",
"url": "https://access.redhat.com/errata/RHSA-2018:0583"
},
{
"type": "WEB",
"url": "https://access.redhat.com/errata/RHSA-2018:0585"
},
{
"type": "PACKAGE",
"url": "https://github.com/rubygems/rubygems"
},
{
"type": "WEB",
"url": "https://lists.debian.org/debian-lts-announce/2018/07/msg00012.html"
},
{
"type": "WEB",
"url": "https://security.gentoo.org/glsa/201710-01"
},
{
"type": "WEB",
"url": "https://web.archive.org/web/20170907215801/http://www.securitytracker.com/id/1039249"
},
{
"type": "WEB",
"url": "https://web.archive.org/web/20170915000000*/http://www.securityfocus.com/bid/100576#:~:text=1%20snapshot-,11%3A49%3A33,-Note"
},
{
"type": "WEB",
"url": "https://www.debian.org/security/2017/dsa-3966"
},
{
"type": "WEB",
"url": "http://blog.rubygems.org/2017/08/27/2.6.13-released.html"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
],
"summary": "RubyGems Code Injection vulnerability"
}
GHSA-7GFH-X38P-PRH3
Vulnerability from github – Published: 2026-07-24 16:21 – Updated: 2026-08-13 18:05Summary
Remote Code Execution (RCE) in velocityjs v2.1.6 via property-read to the Function constructor. This bypasses the fix for GHSA-j658-c2gf-x6pq ("Prototype Pollution in #set path assignment") — that advisory blocked constructor/proto/prototype only in the #set assignment handler (set.cjs), but property read expressions are unfiltered. Any application rendering attacker-controlled Velocity templates is vulnerable to arbitrary code execution on the server.
Details
GHSA-j658-c2gf-x6pq added isBlockedPathKey() to dist/cjs/compile/set.cjs:35-43, which blocks proto, constructor, and prototype keys. However, this check only runs when the #set directive assigns a value — it validates the assignment target path, not the value expression being evaluated. The value expression is evaluated via getReferences() in dist/cjs/compile/references.cjs:16, which calls getAttributes() at line 81. The property access at line 88-89 has no filtering: // references.cjs:81-91 getAttributes(property, baseRef, ast) { if (property.type === "property") { return baseRef[property.id]; // ← NO BLOCK on "constructor", "prototype", etc. } ... } Meanwhile, set.cjs:35-43 properly blocks these keys, but only for the #set target: // set.cjs:35-43 isBlockedPathKey(baseRef, key, isEnd) { if (key === PROTO_KEY) return true; // "proto" if (key === "prototype" && typeof baseRef === "function") return true; return !isEnd && PROTOTYPE_CHAIN_KEYS.has(key) && !hasOwnProperty(baseRef, key); } The exploit chain: 1. $x.constructor → getAttributes() → {}["constructor"] → Object 2. .constructor → getAttributes() → Object["constructor"] → Function 3. ("return process.mainModule.require('child_process').execSync('whoami')") → calls Function(...) → creates a function 4. The #set assigns the result to $f, which is then rendered as $r The #set handler validates $f as the assignment target (which passes — f is not blocked), but never inspects the right-hand expression for prototype chain traversal.
PoC
const velocity = require('velocityjs');
const template = "#set($f=$x.constructor.constructor(\"return process.mainModule.require('child_process').execSync('whoami').toString()\"))#set($r=$f())$r";
console.log(velocity.render(template, { x: {} })); // Output: Verified on velocityjs v2.1.6, Node.js v24.15.0, Windows.
Impact
Type: Remote Code Execution (RCE)
Any application that renders attacker-controlled Velocity templates using velocityjs is vulnerable to full server compromise. The attacker can execute arbitrary shell commands, read environment variables, access cloud credentials, and pivot to internal network resources. The existing advisory GHSA-j658-c2gf-x6pq established the threat model: "any app rendering attacker-controlled templates." This finding demonstrates that the fix was incomplete — the #set blocking was added but property read expressions remain unfiltered, making the issue strictly worse (RCE vs prototype pollution).
Patches
Fixed in velocityjs 2.1.7. The fix was merged in pull request #192 and released as v2.1.7. Users should upgrade to version 2.1.7 or later.
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 2.1.6"
},
"package": {
"ecosystem": "npm",
"name": "velocityjs"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2.1.7"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-73649"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": true,
"github_reviewed_at": "2026-07-24T16:21:22Z",
"nvd_published_at": null,
"severity": "CRITICAL"
},
"details": "### Summary\n\nRemote Code Execution (RCE) in velocityjs v2.1.6 via property-read to the Function constructor. This bypasses the fix for GHSA-j658-c2gf-x6pq (\"Prototype Pollution in #set path assignment\") \u2014 that advisory blocked constructor/__proto__/prototype only in the #set assignment handler (set.cjs), but property read expressions are unfiltered. Any application rendering attacker-controlled Velocity templates is vulnerable to arbitrary code execution on the server.\n\n### Details\n\nGHSA-j658-c2gf-x6pq added isBlockedPathKey() to dist/cjs/compile/set.cjs:35-43, which blocks __proto__, constructor, and prototype keys. However, this check only runs when the #set directive assigns a value \u2014 it validates the assignment target path, not the value expression being evaluated.\nThe value expression is evaluated via getReferences() in dist/cjs/compile/references.cjs:16, which calls getAttributes() at line 81. The property access at line 88-89 has no filtering:\n// references.cjs:81-91\ngetAttributes(property, baseRef, ast) {\n if (property.type === \"property\") {\n return baseRef[property.id]; // \u2190 NO BLOCK on \"constructor\", \"prototype\", etc.\n }\n ...\n}\nMeanwhile, set.cjs:35-43 properly blocks these keys, but only for the #set target:\n// set.cjs:35-43\nisBlockedPathKey(baseRef, key, isEnd) {\n if (key === PROTO_KEY) return true; // \"__proto__\"\n if (key === \"prototype\" \u0026\u0026 typeof baseRef === \"function\") return true;\n return !isEnd \u0026\u0026 PROTOTYPE_CHAIN_KEYS.has(key) \u0026\u0026 !hasOwnProperty(baseRef, key);\n}\nThe exploit chain:\n1. $x.constructor \u2192 getAttributes() \u2192 {}[\"constructor\"] \u2192 Object\n2. .constructor \u2192 getAttributes() \u2192 Object[\"constructor\"] \u2192 Function\n3. (\"return process.mainModule.require(\u0027child_process\u0027).execSync(\u0027whoami\u0027)\") \u2192 calls Function(...) \u2192 creates a function\n4. The #set assigns the result to $f, which is then rendered as $r\nThe #set handler validates $f as the assignment target (which passes \u2014 f is not blocked), but never inspects the right-hand expression for prototype chain traversal.\n\n### PoC\n\nconst velocity = require(\u0027velocityjs\u0027);\n\nconst template = \"#set($f=$x.constructor.constructor(\\\"return process.mainModule.require(\u0027child_process\u0027).execSync(\u0027whoami\u0027).toString()\\\"))#set($r=$f())$r\";\n\nconsole.log(velocity.render(template, { x: {} }));\n// Output: \u003ccurrent OS username\u003e\nVerified on velocityjs v2.1.6, Node.js v24.15.0, Windows.\n\n### Impact\n\nType: Remote Code Execution (RCE)\n\nAny application that renders attacker-controlled Velocity templates using velocityjs is vulnerable to full server compromise. The attacker can execute arbitrary shell commands, read environment variables, access cloud credentials, and pivot to internal network resources.\nThe existing advisory GHSA-j658-c2gf-x6pq established the threat model: \"any app rendering attacker-controlled templates.\" This finding demonstrates that the fix was incomplete \u2014 the #set blocking was added but property read expressions remain unfiltered, making the issue strictly worse (RCE vs prototype pollution).\n\n### Patches\n\nFixed in [velocityjs 2.1.7](https://www.npmjs.com/package/velocityjs/v/2.1.7). The fix was merged in [pull request #192](https://github.com/shepherdwind/velocity.js/pull/192) and released as [v2.1.7](https://github.com/shepherdwind/velocity.js/releases/tag/v2.1.7). Users should upgrade to version 2.1.7 or later.",
"id": "GHSA-7gfh-x38p-prh3",
"modified": "2026-08-13T18:05:38Z",
"published": "2026-07-24T16:21:22Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/shepherdwind/velocity.js/security/advisories/GHSA-7gfh-x38p-prh3"
},
{
"type": "WEB",
"url": "https://github.com/shepherdwind/velocity.js/pull/192"
},
{
"type": "WEB",
"url": "https://github.com/shepherdwind/velocity.js/commit/f8e47a6c4607249b9c967d3a1ced959b4dd64dba"
},
{
"type": "PACKAGE",
"url": "https://github.com/shepherdwind/velocity.js"
},
{
"type": "WEB",
"url": "https://github.com/shepherdwind/velocity.js/releases/tag/v2.1.7"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
],
"summary": "Velocity.js: Remote Code Execution via property-read to Function constructor (bypass of GHSA-j658-c2gf-x6pq fix)"
}
GHSA-7GFQ-F96F-G85J
Vulnerability from github – Published: 2023-08-22 21:30 – Updated: 2024-09-30 16:12An issue in langchain v.0.0.171 allows a remote attacker to execute arbitrary code via the via the a json file to the load_prompt parameter. This is related to __subclasses__ or a template.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "langchain"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "0.0.312"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2023-36281"
],
"database_specific": {
"cwe_ids": [
"CWE-94"
],
"github_reviewed": true,
"github_reviewed_at": "2023-08-23T13:22:25Z",
"nvd_published_at": "2023-08-22T19:16:36Z",
"severity": "CRITICAL"
},
"details": "An issue in langchain v.0.0.171 allows a remote attacker to execute arbitrary code via the via the a json file to the `load_prompt` parameter. This is related to `__subclasses__` or a template.",
"id": "GHSA-7gfq-f96f-g85j",
"modified": "2024-09-30T16:12:36Z",
"published": "2023-08-22T21:30:27Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2023-36281"
},
{
"type": "WEB",
"url": "https://github.com/hwchase17/langchain/issues/4394"
},
{
"type": "WEB",
"url": "https://github.com/langchain-ai/langchain/pull/10252"
},
{
"type": "WEB",
"url": "https://github.com/langchain-ai/langchain/commit/22abeb9f6cc555591bf8e92b5e328e43aa07ff6c"
},
{
"type": "WEB",
"url": "https://aisec.today/LangChain-2e6244a313dd46139c5ef28cbcab9e55"
},
{
"type": "PACKAGE",
"url": "https://github.com/langchain-ai/langchain"
},
{
"type": "WEB",
"url": "https://github.com/langchain-ai/langchain/releases/tag/v0.0.312"
},
{
"type": "WEB",
"url": "https://github.com/pypa/advisory-database/tree/main/vulns/langchain/PYSEC-2023-151.yaml"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"type": "CVSS_V4"
}
],
"summary": "langchain vulnerable to arbitrary code execution"
}
Mitigation
Strategy: Refactoring
Refactor your program so that you do not have to dynamically generate code.
Mitigation
- Run your code in a "jail" or similar sandbox environment that enforces strict boundaries between the process and the operating system. This may effectively restrict which code can be executed by your product.
- Examples include the Unix chroot jail and AppArmor. In general, managed code may provide some protection.
- This may not be a feasible solution, and it only limits the impact to the operating system; the rest of your application may still be subject to compromise.
- Be careful to avoid CWE-243 and other weaknesses related to jails.
Mitigation MIT-5
Strategy: Input Validation
- 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.
- To reduce the likelihood of code injection, use stringent allowlists that limit which constructs are allowed. If you are dynamically constructing code that invokes a function, then verifying that the input is alphanumeric might be insufficient. An attacker might still be able to reference a dangerous function that you did not intend to allow, such as system(), exec(), or exit().
Mitigation
Use dynamic tools and techniques that interact with the product using large test suites with many diverse inputs, such as fuzz testing (fuzzing), robustness testing, and fault injection. The product's operation may slow down, but it should not become unstable, crash, or generate incorrect results.
Mitigation MIT-32
Strategy: Compilation or Build Hardening
Run the code in an environment that performs automatic taint propagation and prevents any command execution that uses tainted variables, such as Perl's "-T" switch. This will force the program to perform validation steps that remove the taint, although you must be careful to correctly validate your inputs so that you do not accidentally mark dangerous inputs as untainted (see CWE-183 and CWE-184).
Mitigation MIT-32
Strategy: Environment Hardening
Run the code in an environment that performs automatic taint propagation and prevents any command execution that uses tainted variables, such as Perl's "-T" switch. This will force the program to perform validation steps that remove the taint, although you must be careful to correctly validate your inputs so that you do not accidentally mark dangerous inputs as untainted (see CWE-183 and CWE-184).
Mitigation
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].
CAPEC-242: Code Injection
An adversary exploits a weakness in input validation on the target to inject new code into that which is currently executing. This differs from code inclusion in that code inclusion involves the addition or replacement of a reference to a code file, which is subsequently loaded by the target and used as part of the code of some application.
CAPEC-35: Leverage Executable Code in Non-Executable Files
An attack of this type exploits a system's trust in configuration and resource files. When the executable loads the resource (such as an image file or configuration file) the attacker has modified the file to either execute malicious code directly or manipulate the target process (e.g. application server) to execute based on the malicious configuration parameters. Since systems are increasingly interrelated mashing up resources from local and remote sources the possibility of this attack occurring is high.
CAPEC-77: Manipulating User-Controlled Variables
This attack targets user controlled variables (DEBUG=1, PHP Globals, and So Forth). An adversary can override variables leveraging user-supplied, untrusted query variables directly used on the application server without any data sanitization. In extreme cases, the adversary can change variables controlling the business logic of the application. For instance, in languages like PHP, a number of poorly set default configurations may allow the user to override variables.