PYSEC-2026-2477 flash-attention contains an insecure deserialization vulnerability in its checkpoint loading mechanism

The flash-attention training framework thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 2025-13-04 contains an insecure deserialization vulnerability CWE-502 in its checkpoint loading mechanism. The loadcheckpoint function in checkpoint.py and the checkpoint loading code in eval.py use...

PYSEC-2026-2654 MLflow allows an unauthenticated remote attacker to read arbitrary files from the server's filesystem

A vulnerability in the createmodelversion handler of mlflow/server/handlers.py in mlflow/mlflow versions 3.9.0 and earlier allows an unauthenticated remote attacker to read arbitrary files from the server's filesystem. The issue arises when a CreateModelVersion request includes the tag...

PYSEC-2026-2601 LiteLLM has a sandbox escape in custom-code guardrail

Impact The POST /guardrails/testcustomcode endpoint runs user-supplied Python inside a hand-rolled sandbox. The sandbox can be escaped using bytecode-level techniques, allowing arbitrary code execution in the proxy process — which runs as root in the default Docker image. Reaching the endpoint...

PYSEC-2026-2737 Open WebUI Arbitrary File Write, Delete via Path Traversal

CONFIDENTIAL Vulnerability Disclosure Analysis Documentation ----------------------------------------------- Vulnerability Details --------------------- 1. Discoverer: Taylor Pennington of KoreLogic, Inc. 2. Date Submitted: June 11, 2024 3. Title: Open WebUI Arbitrary File Write, Delete via Path...

PYSEC-2026-2717 Open WebUI Vulnerable to Arbitrary File Upload and Path Traversal

CONFIDENTIAL KL-CAN-2024-002 Vulnerability Details | | Field | Value | |---|-------|-------| | 1 | Discoverer | Jaggar Henry & Sean Segreti of KoreLogic, Inc. | | 2 | Date Submitted | 2024.03.12 | | 3 | Title | Open WebUI Arbitrary File Upload + Path Traversal | | 5 | Affected Vendor | Open WebUI...

PYSEC-2026-2853 OpenStack Cyborg uses rule:allow (check_str='@') as the default policy for multiple API endpoints

OpenStack Cyborg before 16.0.1 uses rule:allow checkstr='@' as the default policy for multiple API endpoints. This unconditionally authorizes any request carrying a valid Keystone token regardless of roles, project membership, or scope. An authenticated user with zero role assignments can complet...

PYSEC-2026-2537 JupyterLab's command linker attributes in HTML enable one-click command execution from untrusted content

JupyterLab's HTML sanitizer allowlists data-commandlinker-command and data-commandlinker-args on button elements, while CommandLinker listens for all click events on document.body and executes the named command without checking whether the element came from trusted JupyterLab UI. A notebook with ...

PYSEC-2026-2681 JupyterLab's command linker attributes in HTML enable one-click command execution from untrusted content

JupyterLab's HTML sanitizer allowlists data-commandlinker-command and data-commandlinker-args on button elements, while CommandLinker listens for all click events on document.body and executes the named command without checking whether the element came from trusted JupyterLab UI. A notebook with ...

PYSEC-2026-3426 wireshark-mcp vulnerable to arbitrary file write via export_objects when WIRESHARK_MCP_ALLOWED_DIRS is not configured

Description Impact wireshark-mcp exposes a wiresharkexportobjects MCP tool that accepts an attacker-controlled destdir parameter and passes it to tshark's --export-objects flag with no mandatory path restriction. The path sandbox alloweddirs is None by default and only activates when the...

PYSEC-2026-2893 PPTAgent: Arbitrary File Write + Directory Creation via markdown_table_to_image

Summary The markdowntabletoimage tool accepts a caller-controlled path parameter and passes it directly to gethtmltableimage: python pptagent/mcpserver.py:127-143 def markdowntabletoimagemarkdowntable: str, path: str, css: str - str: """ Args: path str: The file path where the image will be saved...

PYSEC-2026-2894 PPTAgent: Arbitrary File Write via `save_generated_slides`

Summary This vulnerability has been fixed in https://github.com/icip-cas/PPTAgent/commit/418491a9a1c02d9d93194b5973bb58df35cf9d00. The savegeneratedslides MCP tool accepts a pptxpath argument and writes the generated PPTX file to that path without any workspace restriction or path validation:...

PYSEC-2026-2892 PPTAgent: Arbitrary Code Execution via Python eval() of LLM-Generated Code with Builtins in Scope

Summary This vulnerability has been fixed in https://github.com/icip-cas/PPTAgent/commit/418491a9a1c02d9d93194b5973bb58df35cf9d00. CodeExecutor.executeactions pptagent/apis.py:126-205 processes LLM-generated slide editing actions using Python's eval: python pptagent/apis.py:184-186 partialfunc =...

PYSEC-2026-2450 django-mdeditor is Missing Authentication for Critical Function

All versions of the package django-mdeditor are vulnerable to Missing Authentication for Critical Function in the image upload endpoint. An attacker can upload malicious files and achieve arbitrary code execution since this endpoint lacks authentication protection and proper sanitisation of file...

PYSEC-2026-2599 LiteLLM: Authenticated command execution via MCP stdio test endpoints

Impact Two endpoints used to preview an MCP server before saving it — POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list — accepted a full server configuration in the request body, including the command, args, and env fields used by the stdio transport. When called with a stdio...

PYSEC-2026-2521 InstructLab vulnerable to Path Traversal

A flaw was found in InstructLab. A local attacker could exploit a path traversal vulnerability in the chat session handler by manipulating the logsdir parameter. This allows the attacker to create new directories and write files to arbitrary locations on the system, potentially leading to...

PYSEC-2026-2947 PraisonAI Vulnerable to RCE via Automatic tools.py Import

PraisonAI automatically imports ./tools.py from the current working directory when launching certain components. This includes call.py, toolresolver.py, and CLI tool-loading paths. A malicious tools.py placed in the process working directory is executed immediately, allowing arbitrary Python code...

PYSEC-2026-2914 PraisonAI Vulnerable to RCE via Automatic tools.py Import

PraisonAI automatically imports ./tools.py from the current working directory when launching certain components. This includes call.py, toolresolver.py, and CLI tool-loading paths. A malicious tools.py placed in the process working directory is executed immediately, allowing arbitrary Python code...

PYSEC-2026-2895 PraisonAI Vulnerable to Implicit Execution of Arbitrary Code via Automatic `tools.py` Loading

PraisonAI automatically loads a file named tools.py from the current working directory to discover and register custom agent tools. This loading process uses importlib.util.specfromfilelocation and immediately executes module-level code via spec.loader.execmodule without explicit user consent,...

PYSEC-2026-2949 PraisonAIAgents: Arbitrary File Read via read_skill_file Missing Workspace Boundary and Approval Gate

Summary readskillfile in skilltools.py allows reading arbitrary files from the filesystem by accepting an unrestricted skillpath parameter. Unlike filetools.readfile which enforces workspace boundary confinement, and unlike runskillscript which requires critical-level approval, readskillfile has...

PYSEC-2026-2917 PraisonAI has Template Injection in Agent Tool Definitions

Summary Direct insertion of unescaped user input into template-rendering tools allows arbitrary code execution via specially crafted agent instructions. Details The createagentcentrictools function returns tools like acpcreatefile that process file content using template rendering. When user inpu...