CVE-2026-48774

ProxySQL is a proxy for MySQL and its forks, as well as PostgreSQL. In versions 3.0.0 through 3.0.8, ProxySQL's GenAI/MCP runsqlreadonly tool violates its documented read-only contract for MySQL targets. The tool validates only the full input string with a substring blacklist and first-keyword...

Malicious code in @briskforge/envcheck (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 09dba573f5d6cb00b09562870f2148b3e539786f5d801f2a263338301d759313 The package advertises itself as a tiny environment-variable validator but ships lib/preflight.js, a heavily obfuscated obfuscator.io string-array...

Astra Linux – Vulnerability in Linux

A flaw was discovered in the way RTAS handled memory accesses in user space for kernel communication. In a guest system that is under lockdown usually due to Secure Boot, running on top of PowerVM or KVM hypervisors pseries platform, a root-like local user could exploit this flaw to further...

PT-2026-50922

Name of the Vulnerable Software and Affected Versions Chromacam version 4.0.3.0 Description An unquoted service path issue exists in the PsyFrameGrabberService. This allows local attackers with write access to C: or subdirectories such as C:Program Files x86Personify to execute arbitrary code. By...

CVE-2026-56075

PraisonAI before 4.5.128 contains an arbitrary shell command execution vulnerability where the UI modules hardcode approvalmode to auto, overriding administrator configuration from PRAISONAPPROVALMODE environment variable. Authenticated attackers can instruct the LLM agent to execute arbitrary...

GHSA-V847-HXXW-3PXG PraisonAI recipe.run_stream skips dangerous-tool policy enforcement

PraisonAI recipe.runstream skips dangerous-tool policy enforcement Summary PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...

PraisonAI recipe.run_stream skips dangerous-tool policy enforcement

PraisonAI recipe.runstream skips dangerous-tool policy enforcement Summary PraisonAI recipe execution blocks default-denied dangerous tools unless the caller explicitly passes allowdangeroustools=True. The normal recipe.run path enforces this with checktoolpolicy. The streaming path,...

Information Exposure

Overview PraisonAI is a PraisonAI is an AI Agents Framework with Self Reflection. PraisonAI application combines PraisonAI Agents, AutoGen, and CrewAI into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customisation, and efficient human-agent...

GHSA-R2XF-7JW5-PJG6 Docker MCP Gateway: Argument injection via OCI image label YAML

Summary A maliciously crafted OCI image label can inject arbitrary arguments into the docker run command line constructed by the MCP Gateway. An attacker who controls an image that the victim references via docker://, or that the victim's catalog pulls a snapshot from, can mount the host...

PT-2026-50746

Name of the Vulnerable Software and Affected Versions Docker MCP Plugin affected versions not specified Description A flaw in the OCI image label parsing allows an attacker to inject arbitrary arguments into the docker run command line. This occurs because the io.docker.server.metadata label is...

GHSA-7P36-FQ2R-4H7R Pimcore CMS Twig Sandbox Bypass via SecurityPolicy checkMethodAllowed

Pimcore CMS/DXP version 12.3.8 contains a sandbox bypass vulnerability that allows authenticated administrative attackers to execute arbitrary methods on PHP objects by exploiting empty checkMethodAllowed and checkPropertyAllowed implementations in the custom Twig SecurityPolicy. Attackers can...

CVE-2026-53873 picklescan - Arbitrary Code Execution via profile.run() Blocklist Bypass

picklescan before 1.0.4 contains an incomplete blocklist for the profile module that fails to block the module-level profile.run function, allowing attackers to achieve arbitrary code execution via exec. Attackers can craft malicious pickle files calling profile.runstatement to execute arbitrary...

The Red Agent POV: How it Reasoned its Way to SSRF

Part 1: How the Red Agent uncovered a multi-step attack chain allowing SSRF-to-Local-File-Read on a GCP Cloud Run API...

Deno: Permission Bypass via Unicode Normalization Mismatch on macOS (APFS)

Summary Deno's permission system enforces filesystem and execution restrictions by comparing the requested path against the path supplied to --deny-read, --deny-write, --deny-run, or --deny-ffi. On macOS, that comparison was done at the raw-byte level while the APFS filesystem treats different...

GHSA-HV7X-3X78-GX53 n8n: Wrong OAuth Scope On Evaluations Test Run Creation Endpoint

Impact The POST /workflows/workflowId/test-runs/new endpoint authorized access using workflow:read rather than workflow:execute. An authenticated user with read-only access to a workflow could trigger a real evaluation test run, causing the workflow to execute via the internal workflow runner. Th...

EUVD-2025-210165

Nokia SR Linux is vulnerable to a local privilege escalation vulnerability. Successful exploitation of this vulnerability may allow an authenticated user to execute arbitrary commands with superuser privilege...

PT-2026-50145

Name of the Vulnerable Software and Affected Versions Deno versions prior to 2.7.14 Description Deno's permission system on macOS enforces filesystem and execution restrictions by comparing requested paths against those supplied to --deny-read, --deny-write, --deny-run, or --deny-ffi. The...

MAL-2026-5798 Malicious code in @resolvx/core (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 4639df1cd39850efb8106cbc5ecf3648f386c0cc5cff6c457d90f6a4d569cef0 On npm install, scripts/postinstall.js connects to a hardcoded attacker IP http://213.218.160.189:8080, fallback:80, sends a base64-encoded host...

Malicious code in npm-sandbox-research-c5d6 (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector e7dd3f64f94b15f73c62c5733a5910802ff22adc514e0eb08e153817fcd4158b The package declares a postinstall hook "postinstall": "node run.js" that executes automatically on npm install. The shipped beacon scripts...

CVE-2026-53831

OpenClaw before 2026.5.18 contains a policy enforcement vulnerability in system.run safe-bin allowlist validation that allows shell expansion to modify command interpretation on POSIX nodes. Authenticated operators can exploit shell metacharacters in approved commands to read unintended node-loca...