39 matches found
CVE-2026-85690
Plandex 2.2.1 is vulnerable to a path traversal in the ApplyFiles function, which allows files to be written outside the project directory. Attackers can influence model output through poisoned repository files or attacker-controlled context to write to sensitive locations such as shell rc or cro...
PT-2026-83668
Loop with Unreachable Exit Condition Infinite Loop vulnerability in ash-project ash ai allows an attacker who can influence a model's output to hang the tool loop and drive unbounded, repeated model requests. AshAi.ToolLoop classifies a model response of :tool calls, then filters the calls throug...
PT-2026-80328
Summary Xinference used Python's unsafe eval function when parsing Llama3 tool-call output generated by a large language model. Because the model output can be influenced by attacker-controlled prompts sent to the chat completion API, a remote attacker can craft prompts that cause the model to...
CVE-2026-19969: Buffer Copy without Checking Size of Input
A security vulnerability has been detected in Open Asset Import Library Assimp 17c12da. The impacted element is the function Assimp::MDLImporter::GenerateOutputMeshes3DGSMDL7 of the file code/AssetLib/MDL/MDLLoader.cpp of the component 3DGS MDL7 Model Output Mesh Generator. The manipulation leads...
CVE-2026-72742 DSPy 3.3.0b1 Local File Read via Image/Audio Output Field Parsing
DSPy 3.3.0b1 contains a file exfiltration vulnerability in the Image and Audio output field adapters that allows attackers with influence over language model outputs to read arbitrary local files by injecting a filesystem path into the url field of a parsed Image or Audio typed output. The...
PT-2026-70872
DSPy 3.3.0b1 contains a file exfiltration vulnerability in the Image and Audio output field adapters that allows attackers with influence over language model outputs to read arbitrary local files by injecting a filesystem path into the url field of a parsed Image or Audio typed output. The...
HijackKV: New Threat in Position-Independent KV Cache Reuse
Key-Value KV cache reduces inference latency in large language models LLMs. Traditional prefix-based reuse has low cache hit rates across inference requests because it requires exact token and position matches. To improve efficiency, recent system optimizations introduce position-independent KV...
CVE-2026-45792
rtk filters and compresses command outputs before they reach your LLM context. Prior to 0.32.0, RTK Rust Token Killer improperly trusts project-local configuration files. RTK automatically loads .rtk/filters.toml from the working directory with highest priority and without user notification. An...
summary-awi-poc
summary-awi-poc Public proof-of-concept repository for valida...
Insertion of Sensitive Information into Log File
Overview langsmith is a Client library to connect to the LangSmith Observability and Evaluation Platform. Affected versions of this package are vulnerable to Insertion of Sensitive Information into Log File through the Client handling of events. An attacker can bypass redaction controls and...
CVE-2026-33873
Langflow is a tool for building and deploying AI-powered agents and workflows. Prior to version 1.9.0, the Agentic Assistant feature in Langflow executes LLM-generated Python code during its validation phase. Although this phase appears intended to validate generated component code, the...
CVE-2026-32097
PingPong is a platform for using large language models LLMs for teaching and learning. Prior to 7.27.2, an authenticated user may be able to retrieve or delete files outside the intended authorization scope. This issue could result in retrieval or deletion of private files, including user-uploade...
CVE-2026-25802
CVE context: The connected GHSA advisory GHSA-299V-8PQ9-5GJQ documents a potential XSS in a new API’s MarkdownRenderer component. The vulnerable path is in MarkdownRenderer.jsx (lines 212–231) that uses dangerouslySetInnerHTML to render model-generated HTML. Impact: potential XSS if the model out...
RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance
Deep neural networks are highly susceptible to backdoor attacks, yet most defense methods to date rely on balanced data, overlooking the pervasive class imbalance in real-world scenarios that can amplify backdoor threats. This paper presents the first in-depth investigation of how the dataset...
CVE-2025-65368
SparkyFitness v0.15.8.2 is vulnerable to Cross Site Scripting XSS via user input and LLM output...
CVE-2025-65368
SparkyFitness v0.15.8.2 is vulnerable to Cross Site Scripting XSS via user input and LLM output...
CVE-2025-65368
SparkyFitness v0.15.8.2 is vulnerable to Cross Site Scripting XSS via user input and LLM output...
CVE-2025-65368
SparkyFitness v0.15.8.2 is vulnerable to Cross Site Scripting XSS via user input and LLM output...
PT-2026-3106
Name of the Vulnerable Software and Affected Versions SparkyFitness version 0.15.8.2 Description SparkyFitness is susceptible to Cross-Site Scripting XSS attacks. The issue stems from improper handling of user input and output from Large Language Models LLMs. This allows for the injection of...