CVE-2026-15583 SSRF (confused deputy) in Grafana MCP Server via X-Grafana-URL header

A confused-deputy flaw in Grafana MCP Server allows an unauthenticated remote attacker to exfiltrate the server's environment-configured Grafana service-account token by supplying a crafted X-Grafana-URL request header. This also enables SSRF against arbitrary internal services, including cloud...

CVE-2026-15583

A confused-deputy flaw in Grafana MCP Server allows an unauthenticated remote attacker to exfiltrate the server's environment-configured Grafana service-account token by supplying a crafted X-Grafana-URL request header. This also enables SSRF against arbitrary internal services, including cloud...

EUVD-2026-44597

A confused-deputy flaw in Grafana MCP Server allows an unauthenticated remote attacker to exfiltrate the server's environment-configured Grafana service-account token by supplying a crafted X-Grafana-URL request header. This also enables SSRF against arbitrary internal services, including cloud...

CVE-2026-15583

Grafana MCP Server is affected by CVE-2026-15583, a confusion-deputy vulnerability that allows an unauthenticated remote attacker to exfiltrate the server’s Grafana service-account token and perform SSRF via a crafted X-Grafana-URL header. The flaw enables targeting internal services, including c...

CVE-2026-30623

LiteLLM version 1.18.10 is vulnerable to remote code execution (RCE) via its MCP server creation functionality. The root cause is a lack of validation for arbitrary command and args values provided in a JSON configuration, which allows attackers to execute unauthorized operating system commands w...

PT-2026-58884

Name of the Vulnerable Software and Affected Versions Grafana MCP Server affected versions not specified Description A confused-deputy flaw allows an unauthenticated remote attacker to exfiltrate the environment-configured Grafana service-account token. This is achieved by supplying a crafted...

CVE-2026-30623

LiteLLM 1.18.10 contains a remote code execution vulnerability in its MCP server creation functionality. The application allows users to add MCP servers via a JSON configuration specifying arbitrary command and args values. LiteLLM executes these values on the host without validation, enabling...

CVE-2026-30623: Improper Neutralization of Special Elements used in a Command

LiteLLM 1.18.10 contains a remote code execution vulnerability in its MCP server creation functionality. The application allows users to add MCP servers via a JSON configuration specifying arbitrary command and args values. LiteLLM executes these values on the host without validation, enabling...

CVE-2026-30623

LiteLLM 1.18.10 contains a remote code execution vulnerability in its MCP server creation functionality. The application allows users to add MCP servers via a JSON configuration specifying arbitrary command and args values. LiteLLM executes these values on the host without validation, enabling...

CVE-2026-49988: Exposure of Sensitive Information to an Unauthorized Actor

Repomix is a tool that packs repositories into AI-friendly files. Prior to 1.14.1, the Repomix MCP server attachpackedoutput and readrepomixoutput flow can register and read arbitrary local .json, .txt, .md, or .xml files without the filesystemreadfile runSecretLint safety check or Repomix...

CVE-2026-15583: Externally Controlled Reference to a Resource in Another Sphere

A confused-deputy flaw in Grafana MCP Server allows an unauthenticated remote attacker to exfiltrate the server's environment-configured Grafana service-account token by supplying a crafted X-Grafana-URL request header. This also enables SSRF against arbitrary internal services, including cloud...

CVE-2026-30623

LiteLLM 1.18.10 contains a remote code execution vulnerability in its MCP server creation functionality. The application allows users to add MCP servers via a JSON configuration specifying arbitrary command and args values. LiteLLM executes these values on the host without validation, enabling...

CVE-2026-15643

AWS HealthLake MCP Server awslabs.healthlake-mcp-server is a Model Context Protocol server that enables AI assistants to interact with AWS HealthLake FHIR datastores. A server-side request forgery in the pagination handling component in AWS awslabs.healthlake-mcp-server before 0.0.14 on all...

CVE-2026-15643 AWS HealthLake MCP Server SSRF via Pagination URL

AWS HealthLake MCP Server awslabs.healthlake-mcp-server is a Model Context Protocol server that enables AI assistants to interact with AWS HealthLake FHIR datastores. A server-side request forgery in the pagination handling component in AWS awslabs.healthlake-mcp-server before 0.0.14 on all...

CVE-2026-15643

CVE-2026-15643 concerns AWS HealthLake MCP Server (awslabs.healthlake-mcp-server) prior to 0.0.14. A server-side request forgery in the pagination handling component can be exploited by a remote authenticated user to exfiltrate AWS temporary security credentials to an attacker-controlled endpoint...

PT-2026-58721

AWS HealthLake MCP Server awslabs.healthlake-mcp-server is a Model Context Protocol server that enables AI assistants to interact with AWS HealthLake FHIR datastores. A server-side request forgery in the pagination handling component in AWS awslabs.healthlake-mcp-server before 0.0.14 on all...

CVE-2026-15643: Server-Side Request Forgery (SSRF)

AWS HealthLake MCP Server awslabs.healthlake-mcp-server is a Model Context Protocol server that enables AI assistants to interact with AWS HealthLake FHIR datastores. A server-side request forgery in the pagination handling component in AWS awslabs.healthlake-mcp-server before 0.0.14 on all...

Path Traversal

Overview @zereight/mcp-gitlab is a GitLab MCP server for projects, merge requests, issues, pipelines, wiki, releases, and more Affected versions of this package are vulnerable to Path Traversal in the job URL construction in build/index.js, which interpolates the jobid parameter into GitLab API...

Cortex has Untrusted Project Bootstrap Code Execution via `CLAUDE_PROJECT_DIR`

Untrusted Project Bootstrap Code Execution via CLAUDEPROJECTDIR SummaryThe Cortex MCP server neuro-cortex-memory treats the CLAUDEPROJECTDIR environment variable — automatically set by Claude Code to the currently open project directory — as a trusted Cortex developer checkout. When the...

PYSEC-2026-2676 Cortex has Untrusted Project Bootstrap Code Execution via `CLAUDE_PROJECT_DIR`

Untrusted Project Bootstrap Code Execution via CLAUDEPROJECTDIR Summary The Cortex MCP server neuro-cortex-memory treats the CLAUDEPROJECTDIR environment variable — automatically set by Claude Code to the currently open project directory — as a trusted Cortex developer checkout. When the...