CVE-2026-32112

ha-mcp is a Home Assistant MCP Server. Prior to 7.0.0, the ha-mcp OAuth consent form renders user-controlled parameters via Python f-strings with no HTML escaping. An attacker who can reach the OAuth endpoint and convince the server operator to follow a crafted authorization URL could execute...

CVE-2026-32111

ha-mcp is a Home Assistant MCP Server. Prior to 7.0.0, the ha-mcp OAuth consent form beta feature accepts a user-supplied haurl and makes a server-side HTTP request to haurl/api/config with no URL validation. An unauthenticated attacker can submit arbitrary URLs to perform internal network...

CVE-2026-33060

CKAN MCP Server is a tool for querying CKAN open data portals. Versions prior to 0.4.85 provide tools including ckanpackagesearch and sparqlquery that accept a baseurl parameter, making HTTP requests to arbitrary endpoints without restriction. A CKAN portal client has no legitimate reason to...

CVE-2026-27826

MCP Atlassian is a Model Context Protocol MCP server for Atlassian products Confluence and Jira. Prior to version 0.17.0, an unauthenticated attacker who can reach the mcp-atlassian HTTP endpoint can force the server process to make outbound HTTP requests to an arbitrary attacker-controlled URL b...

A Bootiful Podcast: Daniel Garnier-Moiroux on MCP Security

Hi Spring, AI, Spring AI, security, and Spring Security fans! In this installment I talk to the legendary Daniel-Garnier Moiroux! ai mcp security java...

@grackle-ai/cli (>=0.19.0, <=0.70.1), @grackle-ai/server (>=0.19.0, <=0.70.1) potentially affected by unknown CVE via @grackle-ai/mcp (>=0.19.0, <=0.70.1)

@grackle-ai/mcp NPM version =0.19.0, =0.19.0, =0.19.0, =0.70.1 Source cves: unknown CVE Source advisory: SNYK:JS-GRACKLEAIMCP-15874101...

@grackle-ai/cli (>=0.19.0, <=0.70.1), @grackle-ai/server (>=0.19.0, <=0.70.1) potentially affected by unknown CVE via @grackle-ai/mcp (>=0.19.0, <=0.70.1)

@grackle-ai/mcp NPM version =0.19.0, =0.19.0, =0.19.0, =0.70.1 Source cves: unknown CVE Source advisory: OSV:GHSA-647H-P824-99W7...

CVE-2026-33252

A flaw was found in the Go MCP SDK's Streamable HTTP transport, which uses Go's standard encoding/json package. In deployments without authorization, a remote attacker can exploit this Cross-Site Request Forgery CSRF vulnerability. By sending browser-generated cross-site POST requests to a local...

PT-2026-34753

Name of the Vulnerable Software and Affected Versions radare2-mcp versions 1.6.0 and earlier Description An OS command injection flaw allows remote, unauthenticated attackers to execute arbitrary commands on the host system. This is achieved by bypassing the command filter using shell...

CVE-2026-33252 MCP Go SDK Allows Cross-Site Tool Execution for HTTP Servers without Authorizatrion

The Go MCP SDK used Go's standard encoding/json. Prior to version 1.4.1, the Go SDK's Streamable HTTP transport accepted browser-generated cross-site POST requests without validating the Origin header and without requiring Content-Type: application/json. In deployments without Authorization,...

EUVD-2026-14643

The Go MCP SDK used Go's standard encoding/json. Prior to version 1.4.1, the Go SDK's Streamable HTTP transport accepted browser-generated cross-site POST requests without validating the Origin header and without requiring Content-Type: application/json. In deployments without Authorization,...

CVE-2026-33252 MCP Go SDK Allows Cross-Site Tool Execution for HTTP Servers without Authorizatrion

The Go MCP SDK used Go's standard encoding/json. Prior to version 1.4.1, the Go SDK's Streamable HTTP transport accepted browser-generated cross-site POST requests without validating the Origin header and without requiring Content-Type: application/json. In deployments without Authorization,...

CVE-2026-33252 MCP Go SDK Allows Cross-Site Tool Execution for HTTP Servers without Authorizatrion

The Go MCP SDK used Go's standard encoding/json. Prior to version 1.4.1, the Go SDK's Streamable HTTP transport accepted browser-generated cross-site POST requests without validating the Origin header and without requiring Content-Type: application/json. In deployments without Authorization,...

CVE-2026-23882

Blinko (AI-powered card note-taking) before version 1.8.4 is affected by a remote command execution in the MCP (Model Context Protocol) server creation function, which allowed specifying arbitrary commands and arguments that are executed during connection testing. The issue is patched in version ...

CVE-2026-23882 Blinko: Admin RCE - MCP Server Command Injection

Blinko is an AI-powered card note-taking project. Prior to version 1.8.4, the MCP Model Context Protocol server creation function allows specifying arbitrary commands and arguments, which are executed when testing the connection. This issue has been patched in version 1.8.4...

EUVD-2026-14473

A flaw has been found in erupts erupt bis 1.13.3. Affected by this vulnerability is the function EruptDataQuery of the file erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java of the component MCP Tool Interface. This manipulation causes sql injection hibernate. It is possible to...

CVE-2026-4593 erupts erupt MCP Tool EruptDataQuery.java EruptDataQuery sql injection

A flaw has been found in erupts erupt bis 1.13.3. Affected by this vulnerability is the function EruptDataQuery of the file erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java of the component MCP Tool Interface. This manipulation causes sql injection hibernate. It is possible to...

DNS Rebinding

@modelcontextprotocol/sdk is vulnerable to DNS Rebinding. The vulnerability is due to DNS rebinding protection being disabled by default in unauthenticated HTTP-based servers, which allows an attacker to exploit a malicious website to bypass same-origin policy and send requests to the local MCP...

Command Injection

MCP Watch is vulnerable to Command Injection. The vulnerability is due to unsanitized user input being passed to execSync in the cloneRepo method, which allows an attacker to append shell metacharacters to the URL and execute arbitrary commands on the host system...

Auditing MCP Servers for Over-Privileged Tool Capabilities

The Model Context Protocol MCP has emerged as a standard for connecting Large Language Models LLMs to external tools and data. However, MCP servers often expose privileged capabilities, such as file system access, network requests, and command execution that can be exploited if not properly...