GHSA-VJ5C-M527-MPFF
Vulnerability from github – Published: 2026-07-28 22:23 – Updated: 2026-07-28 22:23Impact
Prototype pollution.
A malicious user can create a token array [{ key: '{__proto__.foo}', value: 'malicious' }], when processed by convertTokenData() utility function, it will pollute the Object.prototype globally where {}.foo will equal { key: '{__proto__.foo}', value: 'malicious' }.
This has been confirmed with a test/reproduction.
You are impacted when:
- direct usage of convertTokenData(tokens, { output: 'object' });
- indirect usage, via using Expand API https://styledictionary.com/reference/config/#expand. If your expand config deems it necessary to run expand (this means, if NOT: 1) set to false, 2) all subprops set to false, or 3) undefined), then we sync the sd.tokens property with the sd.tokenMap property by converting tokenData map back to object.
- indirect usage via SD's transform lifecycle. Once your tokens are transformed, we also have to sync the sd.tokens property with the sd.tokenMap property.
Impact is high for this when style-dictionary is used as an integration in a NodeJS server application. Impact is moderate for when style-dictionary is used as an integration in a Web application. Impact is low for most common cases where the user of style-dictionary also maintains the tokens, and access is limited via read/write access to the repository/workflows where it is used.
Patches
A patch has been published: version 5.4.4.
Any version within range >=4.3.0 <5.4.4 contains this vulnerability, see commit hash 209085d for when the vulnerability was added.
See PR with repro + fix https://github.com/style-dictionary/style-dictionary/pull/1702
Workarounds
A workaround is to sanitize your token data first. Whether using DTCG format or old Style Dictionary format, you have to check the token data object recursively for any object keys that include __proto__.
You can do this with the StyleDictionary instance too, just ensure that expand has to be set to false to prevent the second method of this vulnerability from happening.
const sd = new StyleDictionary({ expand: false });
if (sd.allTokens.some(tok => tok.key.includes('__proto__')) {
throw new Error('Found malicious token key, attempting to do prototype pollution.')
}
{
"affected": [
{
"package": {
"ecosystem": "npm",
"name": "style-dictionary"
},
"ranges": [
{
"events": [
{
"introduced": "4.3.0"
},
{
"fixed": "5.4.4"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-54639"
],
"database_specific": {
"cwe_ids": [
"CWE-1321"
],
"github_reviewed": true,
"github_reviewed_at": "2026-07-28T22:23:03Z",
"nvd_published_at": "2026-06-24T01:16:25Z",
"severity": "HIGH"
},
"details": "### Impact\nPrototype pollution.\nA malicious user can create a token array `[{ key: \u0027{__proto__.foo}\u0027, value: \u0027malicious\u0027 }]`, when processed by `convertTokenData()` utility function, it will pollute the Object.prototype globally where `{}.foo` will equal `{ key: \u0027{__proto__.foo}\u0027, value: \u0027malicious\u0027 }`.\n\nThis has been confirmed with a test/reproduction.\n\nYou are impacted when:\n- direct usage of `convertTokenData(tokens, { output: \u0027object\u0027 });`\n- indirect usage, via using Expand API https://styledictionary.com/reference/config/#expand. If your expand config deems it necessary to run expand (this means, if NOT: 1) set to false, 2) all subprops set to false, or 3) undefined), then we sync the `sd.tokens` property with the `sd.tokenMap` property by converting tokenData map back to object.\n- indirect usage via SD\u0027s transform lifecycle. Once your tokens are transformed, we also have to sync the `sd.tokens` property with the `sd.tokenMap` property.\n\nImpact is high for this when style-dictionary is used as an integration in a NodeJS server application.\nImpact is moderate for when style-dictionary is used as an integration in a Web application.\nImpact is low for most common cases where the user of style-dictionary also maintains the tokens, and access is limited via read/write access to the repository/workflows where it is used.\n\n### Patches\nA patch has been published: version `5.4.4`.\nAny version within range `\u003e=4.3.0 \u003c5.4.4` contains this vulnerability, see commit hash 209085d for when the vulnerability was added.\n\nSee PR with repro + fix https://github.com/style-dictionary/style-dictionary/pull/1702\n\n### Workarounds\nA workaround is to sanitize your token data first. Whether using DTCG format or old Style Dictionary format, you have to check the token data object recursively for any object keys that include `__proto__`.\n\nYou can do this with the StyleDictionary instance too, just ensure that expand has to be set to false to prevent the second method of this vulnerability from happening.\n\n```js\nconst sd = new StyleDictionary({ expand: false });\n\nif (sd.allTokens.some(tok =\u003e tok.key.includes(\u0027__proto__\u0027)) {\n throw new Error(\u0027Found malicious token key, attempting to do prototype pollution.\u0027)\n}\n```",
"id": "GHSA-vj5c-m527-mpff",
"modified": "2026-07-28T22:23:03Z",
"published": "2026-07-28T22:23:03Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/style-dictionary/style-dictionary/security/advisories/GHSA-vj5c-m527-mpff"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-54639"
},
{
"type": "WEB",
"url": "https://github.com/style-dictionary/style-dictionary/pull/1702"
},
{
"type": "WEB",
"url": "https://github.com/style-dictionary/style-dictionary/commit/209085d9782cfc0783c4d983f3f1bb2c515954ec"
},
{
"type": "WEB",
"url": "https://github.com/style-dictionary/style-dictionary/commit/23b5e8dda143441f0d6b8e2b4222e2da98058bc5"
},
{
"type": "PACKAGE",
"url": "https://github.com/style-dictionary/style-dictionary"
},
{
"type": "WEB",
"url": "https://github.com/style-dictionary/style-dictionary/releases/tag/v5.4.4"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
"type": "CVSS_V3"
}
],
"summary": "Style Dictionary - Prototype Pollution in convertTokenData utility function"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.