CVE-2026-7669

Published: Mag 03, 2026 Last Modified: Mag 04, 2026
ExploitDB:
Other exploit source:
Google Dorks:
MEDIUM 6,3
Attack Vector: network
Attack Complexity: high
Privileges Required: none
User Interaction: none
Confidentiality: N/A
Integrity: N/A
Availability: N/A
MEDIUM 5,6
Attack Vector: network
Attack Complexity: high
Privileges Required: none
User Interaction: none
Scope: unchanged
Confidentiality: low
Integrity: low
Availability: low
MEDIUM 5,1
Access Vector: network
Access Complexity: high
Authentication: none
Confidentiality: partial
Integrity: partial
Availability: partial

Description

AI Translation Available

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 of the argument trust_remote_code with the input False as part of Boolean results in code injection. The attack can be executed remotely. A high complexity level is associated with this attack. The exploitability is considered difficult. In get_tokenizer(), when the caller passes trust_remote_code=False and HuggingFace transformers v5 returns a TokenizersBackend instance (the generic fallback for tokenizer classes not in the registry), SGLang silently re-invokes AutoTokenizer.from_pretrained with trust_remote_code=True, overriding the caller's explicit security setting. A model repository containing a malicious tokenizer.py referenced via auto_map in tokenizer_config.json will execute arbitrary Python in the SGLang process during this second call. No log line or warning is emitted. The override affects all current SGLang versions because transformers==5.3.0 is pinned in pyproject.toml. Both tokenizer_mode='auto' and tokenizer_mode='slow' are affected. The exploit is now public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.

EPSS (Exploit Prediction Scoring System)

Trend Analysis

EPSS (Exploit Prediction Scoring System)

Prevede la probabilità di sfruttamento basata su intelligence sulle minacce e sulle caratteristiche della vulnerabilità.

EPSS Score
0,0004
Percentile
0,1th
Updated

Single Data Point

Only one EPSS measurement is available for this CVE. Trend analysis requires multiple data points over time.

74

Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')

Incomplete
Common Consequences
Security Scopes Affected:
Confidentiality Access Control Other Integrity Non-Repudiation
Potential Impacts:
Read Application Data Bypass Protection Mechanism Alter Execution Logic Other Hide Activities
Applicable Platforms
All platforms may be affected
View CWE Details
94

Improper Control of Generation of Code ('Code Injection')

Draft
Common Consequences
Security Scopes Affected:
Access Control Integrity Confidentiality Availability Non-Repudiation
Potential Impacts:
Bypass Protection Mechanism Gain Privileges Or Assume Identity Execute Unauthorized Code Or Commands Hide Activities
Applicable Platforms
Languages: Interpreted
Technologies: AI/ML
View CWE Details
https://github.com/gouldnicholas/CVE-2026-7669-PoC
https://vuldb.com/submit/799263
https://vuldb.com/vuln/360817
https://vuldb.com/vuln/360817/cti