These pages are published after PatchSiren validates generated defensive summaries against stored public CVE and source evidence.
Hugging Face Text Embeddings Inference through 1.9.4 contains a cleartext logging vulnerability that exposes the configured api_key because the router's Args struct lacks a redact attribute for it. Attackers with access to router logs, container output, or OTLP telemetry can recover the Bearer token and call the protected embedding and rerank endpoints.
A vulnerability in Hugging Face Transformers (versions 4.57.0 to 5.16.1) allows remote Python files to be written to local disk without user consent when using GenerativePreTrainedModel.load_custom_generate(). The function fetches and caches a remote module file before performing the required trust_remote_code consent check, inverting the security model enforced by other code-loading paths.
AI-assisted PatchSiren debrief based on the supplied source corpus. The CVE record was published on 2026-08-20T17:17:21.050Z and has not been modified since then. This vulnerability, identified as CVE-2026-15679, affects Hugging Face PyTorch Image Models, allowing remote attackers to execute arbitrary code. The vulnerability exists within the parsing of checkpoints in Hugging Face PyTorch Image Models, re [truncated]
CVE-2026-69112 is a path traversal vulnerability in Hugging Face Accelerate versions through 1.14.0. The load_checkpoint_in_model and load_checkpoint_and_dispatch functions do not properly sanitize weight_map entries from sharded checkpoint indexes. This allows attackers to supply relative paths with ../ sequences or absolute paths to read arbitrary files. Additionally, pointing shard entries at named pip [truncated]
AI-assisted PatchSiren debrief based on the supplied source corpus. The CVE record was published on 2026-07-31T21:17:32.440Z and has not been modified since then. The sentence-transformers library contains a security control bypass vulnerability that allows attackers to achieve arbitrary code execution by exploiting a logic flaw in the import_module_class helper within sentence_transformers/util/misc.py. [truncated]
LeRobot through 0.5.1 contains an unsafe deserialization vulnerability in the async inference pipeline where pickle.loads() is used to deserialize data received over unauthenticated gRPC channels without TLS in the policy server and robot client components. This vulnerability allows an unauthenticated network-reachable attacker to achieve arbitrary code execution on the server or client by sending a craft [truncated]