15836 matches found
Stealing Reasoning Traces from Proprietary LLM APIs
Leading large language model providers now conceal their models' step-by-step reasoning, or chain-of-thought, to protect intellectual property and limit information leakage. Rather than storing these traces server-side, providers return them to the client as blocks of encrypted text, which the...
PT-2026-69334
Name of the Vulnerable Software and Affected Versions keras-team/keras versions prior to 3.15.1 Description A denial of service DoS issue occurs when loading malicious .keras model files using the keras.models.load model function. The H5IOStore. getitem method in keras/src/saving/saving lib.py...
PT-2026-76476
Summary Open WebUI lets a client define a model inline on a chat request instead of selecting a saved workspace model. The knowledge attached to such an inline model was used as-is, without checking that the caller can read what it points at. Any authenticated user who knows another user's file i...
CVE-2026-13717: Improper Access Control
A flaw was found in the Red Hat OpenShift AI RHOAI MaaS Gateway. Improper configuration of the Gateway in a model-serving context allows a standard user with low privileges to intercept, read, log, and alter all MaaS model traffic. This includes sensitive information such as access keys, input...
mcp-injection-poc
MCP Transparent Proxy Injection — Proof of Concept CVE: pe...
CVE-2026-19355 MingSoft MCMS ms-mdiy list.do ModelDataImpl.queryDiyFormData sql injection
A vulnerability was determined in MingSoft MCMS up to 3.0.6. This affects the function ModelDataImpl.queryDiyFormData of the file /mdiy/form/data/list.do of the component ms-mdiy. Executing a manipulation of the argument formFields can lead to sql injection. The attack may be performed from remot...
CVE-2026-19355
A vulnerability was determined in MingSoft MCMS up to 3.0.6. This affects the function ModelDataImpl.queryDiyFormData of the file /mdiy/form/data/list.do of the component ms-mdiy. Executing a manipulation of the argument formFields can lead to sql injection. The attack may be performed from remot...
CVE-2026-19355 MingSoft MCMS ms-mdiy list.do ModelDataImpl.queryDiyFormData sql injection
A vulnerability was determined in MingSoft MCMS up to 3.0.6. This affects the function ModelDataImpl.queryDiyFormData of the file /mdiy/form/data/list.do of the component ms-mdiy. Executing a manipulation of the argument formFields can lead to sql injection. The attack may be performed from remot...
MINI-87G3-96GW-6W6W
Bulletin has no description...
Why Generative AI Needs a New Kind of Security
This is Post 2 of a four-part series: Post 1: Agentic AI Security: The Chatbot Era Is Over Post 3: The OWASP LLM Top 10 Was the Warm-Up: What Comes Next Post 4: OWASP LLM Top 10 2026: Every Move Points the Same Direction There is a moment every CISO is having right now, often without warning. A...
CVE-2026-19334
A flaw has been found in NightTrek Ollama-mcp up to 80cf2e17cfc144963a475b619093a2d13c13dbc9. This affects an unknown part of the file src/index.ts. This manipulation of the argument name/modelfile/source/destination causes command injection. The attack can only be executed locally. This product ...
PT-2026-69252
Name of the Vulnerable Software and Affected Versions andreahaku codex mcp versions up to 1ff521cc6cc57cfe56ddef946c644b8534771390 Description Command injection is possible in the ask MCP Tool component within the file src/codex-process-simple.ts. The issue occurs when the model argument is...
Telemetry and Concealment in Self-Adapting Generative AI: Logging Architecture, Adversarial Model Hiding, and the Limits of Detection
Model risk management MRM guidance assumes a static model lifecycle, in which models are developed, independently validated, and implemented without further autonomous modification. Continually self-adapting generative AI systems --- models that update their own weights during production deployme...
EUVD-2026-54805
D-Link DWR-M961 devices with hardware version C1 and firmware version before 1.1.5C1202607071108 contain a command injection vulnerability in the /boafrm/formNtp interface. A remote attacker can inject arbitrary malicious commands into the ntpServerIp1 field, resulting in command execution with...
MINI-733M-XGPR-Q969
Bulletin has no description...
MINI-M8XG-HWFM-R896
Bulletin has no description...
MINI-6X65-G7GV-F759
Bulletin has no description...
CVE-2026-47663
Pathling is a set of tools that make it easier to use FHIR and clinical terminology within health data analytics. Prior to version 2.0.0 of Pathling Server, Pathling's typed CRUD/search/batch FHIR surface allows an authenticated caller with only coarse operation authorities to act on...
CVE-2026-61808
LightRAG provides simple and fast retrieval-augmented generation. Through version 1.5.4, the LightRAG API server binds to all network interfaces with authentication disabled by default, allowing an unauthenticated network attacker to read indexed document content, upload or delete documents, modi...