PYSEC-2026-3062
Vulnerability from pysec - Published: 2026-07-13 15:02 - Updated: 2026-07-13 16:06A vulnerability was detected in sgl-project SGLang up to 0.5.9. Impacted is the function get_tokenizer of the file python/sglang/srt/utils/hf_transformers_utils.py of the component HuggingFace Transformer Handler. The manipulation results in deserialization. The attack can be executed remotely. A high complexity level is associated with this attack. The exploitability is considered difficult. The vendor was contacted early about this disclosure but did not respond in any way.
| Name | purl | sglang | pkg:pypi/sglang |
|---|
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "sglang",
"purl": "pkg:pypi/sglang"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "0.5.9"
}
],
"type": "ECOSYSTEM"
}
],
"versions": [
"0.1.10",
"0.1.11",
"0.1.12",
"0.1.13",
"0.1.14",
"0.1.15",
"0.1.16",
"0.1.17",
"0.1.18",
"0.1.19",
"0.1.20",
"0.1.21",
"0.1.22",
"0.1.24",
"0.1.25",
"0.1.26",
"0.1.3",
"0.1.4",
"0.1.5",
"0.1.6",
"0.1.7",
"0.1.9",
"0.2.0",
"0.2.1",
"0.2.10",
"0.2.11",
"0.2.12",
"0.2.13",
"0.2.14",
"0.2.14.post1",
"0.2.14.post2",
"0.2.15",
"0.2.2",
"0.2.3",
"0.2.4",
"0.2.5",
"0.2.6",
"0.2.7",
"0.2.8",
"0.2.9",
"0.2.9.post1",
"0.3.0",
"0.3.1",
"0.3.1.post1",
"0.3.1.post2",
"0.3.1.post3",
"0.3.2",
"0.3.3",
"0.3.3.post1",
"0.3.4",
"0.3.4.post1",
"0.3.4.post2",
"0.3.5",
"0.3.5.post1",
"0.3.5.post2",
"0.3.6",
"0.3.6.post1",
"0.3.6.post2",
"0.3.6.post3",
"0.4.0",
"0.4.0.post1",
"0.4.0.post2",
"0.4.1",
"0.4.1.post1",
"0.4.1.post2",
"0.4.1.post3",
"0.4.1.post4",
"0.4.1.post5",
"0.4.1.post6",
"0.4.1.post7",
"0.4.10",
"0.4.10.post1",
"0.4.10.post2",
"0.4.2",
"0.4.2.post1",
"0.4.2.post2",
"0.4.2.post3",
"0.4.2.post4",
"0.4.3",
"0.4.3.post1",
"0.4.3.post2",
"0.4.3.post3",
"0.4.3.post4",
"0.4.4",
"0.4.4.post1",
"0.4.4.post2",
"0.4.4.post3",
"0.4.4.post4",
"0.4.5",
"0.4.5.post1",
"0.4.5.post2",
"0.4.5.post3",
"0.4.6",
"0.4.6.post1",
"0.4.6.post2",
"0.4.6.post3",
"0.4.6.post4",
"0.4.6.post5",
"0.4.7",
"0.4.7.post1",
"0.4.8",
"0.4.8.post1",
"0.4.9",
"0.4.9.post1",
"0.4.9.post2",
"0.4.9.post3",
"0.4.9.post4",
"0.4.9.post5",
"0.4.9.post6",
"0.5.0rc0",
"0.5.0rc1",
"0.5.0rc2",
"0.5.1",
"0.5.1.post1",
"0.5.1.post2",
"0.5.1.post3",
"0.5.2",
"0.5.2rc0",
"0.5.2rc1",
"0.5.2rc2",
"0.5.3",
"0.5.3.post1",
"0.5.3.post2",
"0.5.3.post3",
"0.5.3rc0",
"0.5.3rc2",
"0.5.4",
"0.5.4.post1",
"0.5.4.post2",
"0.5.4.post3",
"0.5.5",
"0.5.5.post1",
"0.5.5.post2",
"0.5.5.post3",
"0.5.6",
"0.5.6.post1",
"0.5.6.post2",
"0.5.7",
"0.5.8",
"0.5.8.post1",
"0.5.9"
]
}
],
"aliases": [
"CVE-2026-7669",
"GHSA-6m5f-673f-5vh7"
],
"details": "A vulnerability was detected in sgl-project SGLang up to 0.5.9. Impacted is the function get_tokenizer of the file python/sglang/srt/utils/hf_transformers_utils.py of the component HuggingFace Transformer Handler. The manipulation results in deserialization. The attack can be executed remotely. A high complexity level is associated with this attack. The exploitability is considered difficult. The vendor was contacted early about this disclosure but did not respond in any way.",
"id": "PYSEC-2026-3062",
"modified": "2026-07-13T16:06:17.328054Z",
"published": "2026-07-13T15:02:56.168010Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-7669"
},
{
"type": "WEB",
"url": "https://github.com/gouldnicholas/CVE-2026-7669-PoC"
},
{
"type": "PACKAGE",
"url": "https://github.com/sgl-project/sglang"
},
{
"type": "WEB",
"url": "https://vuldb.com/submit/799263"
},
{
"type": "WEB",
"url": "https://vuldb.com/vuln/360817"
},
{
"type": "WEB",
"url": "https://vuldb.com/vuln/360817/cti"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/sglang"
},
{
"type": "ADVISORY",
"url": "https://github.com/advisories/GHSA-6m5f-673f-5vh7"
}
],
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:L",
"type": "CVSS_V3"
},
{
"score": "CVSS:4.0/AV:N/AC:H/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N",
"type": "CVSS_V4"
}
],
"summary": "SGLang has an Improper Input Validation/Injection Issue"
}
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.
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.