EUVD-2026-61107

NVIDIA Triton Inference Server for Linux contains a vulnerability where an attacker could cause an absolute path traversal. A successful exploit might lead to code execution and information disclosure...

CVE-2026-47606

NVIDIA Triton Inference Server for Linux contains a vulnerability where an attacker could cause an absolute path traversal. A successful exploit might lead to code execution and information disclosure...

CVE-2026-75130

Context7 through 2.1.2 contains a prompt injection vulnerability that allows attackers to execute malicious instructions in connected AI coding agents by injecting unsanitized content through the Custom AI Instructions feature served via the MCP server. Attackers can poison the custom instruction...

CVE-2026-71880

Interpretation of untrusted input in template engine in GBIF Integrated Publishing Toolkit versions before 3.3.4 allows remote authenticated attackers to access server-side files and state via template injection...

CVE-2026-63642

MagicMirror² is an open source modular smart mirror platform. Prior to 2.37.0, checkArticleUrl in defaultmodules/newsfeed/nodehelper.js accepts the CHECKARTICLEURL notification through the unauthenticated Socket.IO namespace /newsfeed and performs fetchurl, method: "HEAD" without validating the...

CVE-2026-63643

MagicMirror² is an open source modular smart mirror platform. Prior to 2.37.0, the ADDCALENDAR handler in defaultmodules/calendar/nodehelper.js accepts an attacker-controlled URL, authentication data, and selfSignedCert setting through the unauthenticated Socket.IO namespace /calendar. The handle...

CVE-2026-63641

MagicMirror² is an open source modular smart mirror platform. Prior to 2.37.0, MagicMirror applies ipWhitelist only as Express middleware, while the Socket.IO server in js/server.js is attached directly to the HTTP server without equivalent IP allowlist, origin, or namespace authentication checks...

CVE-2026-50143

The Apify MCP server enables AI agents to extract data from websites using ready-made scrapers, crawlers, and automation tools available on the Apify Store. Prior to 0.10.11, getActorMCPServerURL in src/mcp/actors.ts concatenates the trusted Actor standby URL with the attacker-controlled...

CVE-2026-24185

NVIDIA NVOS for network switches contains a vulnerability in the secure shell SSH server configuration component while PKA-only mode is enabled, where an administrator could inadvertently enable an alternative authentication path. If best practices for replacing the default password as recommende...

EUVD-2026-61099

Context7 through 2.1.2 contains a prompt injection vulnerability that allows attackers to execute malicious instructions in connected AI coding agents by injecting unsanitized content through the Custom AI Instructions feature served via the MCP server. Attackers can poison the custom instruction...

CVE-2026-75130

Context7 through 2.1.2 has a prompt injection flaw in its Custom AI Instructions feature served via the MCP server. Attackers poison unsanitized instructions to exfiltrate credentials from environment files and delete files on victim machines when the agent makes routine documentation requests. C...

CVE-2026-75130 Context7 2.1.2 Prompt Injection via Custom AI Instructions

Context7 through 2.1.2 contains a prompt injection vulnerability that allows attackers to execute malicious instructions in connected AI coding agents by injecting unsanitized content through the Custom AI Instructions feature served via the MCP server. Attackers can poison the custom instruction...

CVE-2026-49452 WeasyPrint: CSS Injection via Presentational Hints

WeasyPrint helps web developers to create PDF documents. Prior to 69.0, WeasyPrint embeds unescaped HTML presentational-hint attribute values into CSS in weasyprint/css/init.py when presentationalhints=True. The background attribute is inserted into a background-image:url declaration and parsed b...

CVE-2026-71880

GBIF Integrated Publishing Toolkit (versions before 3.3.4) has a server-side template injection vulnerability in its template engine. Remote authenticated attackers with low privileges can inject untrusted input into templates, achieving high confidentiality and integrity impact (CVSS 7.6) by rea...

CVE-2026-71880 Server-side template injection in Integrated Publishing Toolkit

Interpretation of untrusted input in template engine in GBIF Integrated Publishing Toolkit versions before 3.3.4 allows remote authenticated attackers to access server-side files and state via template injection...

CVE-2026-71880 Server-side template injection in Integrated Publishing Toolkit

Interpretation of untrusted input in template engine in GBIF Integrated Publishing Toolkit versions before 3.3.4 allows remote authenticated attackers to access server-side files and state via template injection...

EUVD-2026-61140

Interpretation of untrusted input in template engine in GBIF Integrated Publishing Toolkit versions before 3.3.4 allows remote authenticated attackers to access server-side files and state via template injection...

CVE-2026-50161 libre: Integer overflow in websock_decode() masked frame length check leads to heap buffer overflow

libre is a generic library for real-time communications with asynchronous input and output support. Prior to 4.8.1, the websockdecode function in src/websock/websock.c contains an integer overflow when validating a masked WebSocket frame that uses the 64-bit extended length encoding. The expressi...

CVE-2026-50143

In @apify/actors-mcp-server versions before 0.10.11, getActorMCPServerURL() in src/mcp/actors.ts:44 naively concatenates a trusted base URL with an attacker-controlled webServerMcpPath, allowing an @ character to inject an arbitrary hostname. Node.js's WHATWG URL parser treats the prefix as useri...

CVE-2026-50143 Actor MCP path authority injection leaks Apify token

The Apify MCP server enables AI agents to extract data from websites using ready-made scrapers, crawlers, and automation tools available on the Apify Store. Prior to 0.10.11, getActorMCPServerURL in src/mcp/actors.ts concatenates the trusted Actor standby URL with the attacker-controlled...