CVE-2026-5817-PoC

CVE-2026-5817：Docker Model Runner 容器到宿主机 RCE / 逃逸 在 Docker Desktop（4.40.0 至 4.67.x）宿主机上的任何容器，都可以通过两个 HTTP 请求在宿主机上运行代码。无需套接字挂载，无需 --privileged，无需 capabilities。 原理 每个容器都可以无认证地访问位于 model-runner.docker.internal 的 Model Runner。它会从你指定的任何 OCI 仓库拉取模型，并在不校验摘要的情况下存储它们。Python 后端（vLLM、MLX、SGLang）使用...

CVE-2026-5843

CVE-2026-5843 PoC 一个极简的 OCI 注册表，用于提供恶意模型，以利用 mlx-lm 中的 modelfile importlib 执行路径。当 Docker Model Runner 加载模型进行推理时，model.py 将以当前登录用户的身份在宿主机上执行。 要求： Docker Desktop ≤ 4.70.x（Apple Silicon）并启用 Model Runner 使用方法 root@kitploit: 1. Start the registry on the host python3 poccve20265843.py 2. From any...

The vulnerability of the vllm-metal inference backend component of the Docker Desktop Docker Model Runner plugin allows a attacker to execute arbitrary code.

The vulnerability of the vllm-metal inference backend component of the Docker Desktop Docker Model Runner plugin is related to the inclusion of functions from an unverified controlled area. Exploiting this vulnerability could allow a attacker to execute arbitrary code...

CVE-2026-5817

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5843

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843 Docker Model Runner container-to-host code execution via MLX-LM model_file importlib loading

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843

The CVE describes a vulnerability in Docker Model Runner (macOS) where the MLX-LM backend unconditionally imports and executes arbitrary Python files specified by model_file in a model's config.json via importlib, without a trust_remote_code gate or sandboxing. This enables container-to-host arbi...

EUVD-2026-31491

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5843 Docker Model Runner container-to-host code execution via MLX-LM model_file importlib loading

The MLX inference backend in Docker Model Runner on macOS uses the MLX-LM library, which unconditionally imports and executes arbitrary Python files from model directories via the modelfile configuration field in config.json. When a model's config.json specifies a modelfile pointing to a Python...

CVE-2026-5817 Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

EUVD-2026-31493

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5817

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5817 Docker Model Runner container-to-host code execution via unsandboxed trust_remote_code in Python inference backends

The vllm-metal inference backend in Docker Model Runner on macOS unconditionally sets trustremotecode=True when loading model tokenizers, and runs without sandboxing. This causes transformers.AutoTokenizer.frompretrained to import and execute arbitrary Python files included in any model pulled fr...

CVE-2026-5817

CVE-2026-5817 affects the vllm-metal backend used by Docker Model Runner on macOS. The backend loads model tokenizers with trust_remote_code=True, causing transformers.AutoTokenizer.from_pretrained() to import and execute arbitrary Python files from models pulled from an OCI registry. This can en...

Exploit for CVE-2026-5817

CVE-2026-5817: Docker Model Runner container-to-host RCE / Esc...

PT-2026-42831

Name of the Vulnerable Software and Affected Versions Docker Model Runner on macOS affected versions not specified Description The MLX inference backend uses the MLX-LM library, which imports and executes arbitrary Python files from model directories via the model file configuration field in the...

Docker Model Runner 安全漏洞

Docker Model Runner is an open-source Docker model runner developed by Docker. There is a security vulnerability in Docker Model Runner MLX. This vulnerability stems from the unconditional import and execution of any Python file in the model directory. It may allow malicious models to be pulled...

PT-2026-42830

Name of the Vulnerable Software and Affected Versions Docker Model Runner on macOS affected versions not specified Description The vllm-metal inference backend unconditionally sets trust remote code=True when loading model tokenizers and operates without sandboxing. This allows the...

Docker Model Runner 安全漏洞

Docker Model Runner is an open-source Docker model runner developed by Docker. Docker Model Runner vllm-metal contains a security vulnerability. This vulnerability arises from setting trustremotecode=True without any sandbox protection. It may allow arbitrary Python files to be executed during...