Microsoft & Anthropic MCP Servers at Risk of RCE, Cloud Takeovers
ID: 88256a09-93d8-50b5-994a-f30818c91516
STIX ID: report--88256a09-93d8-50b5-994a-f30818c91516
Feed Name: Dark Reading
Threat Score
Researchers found severe security flaws in popular Model Context Protocol (MCP) servers: a widespread server-side request forgery (SSRF) issue in Microsoft's MarkItDown that can expose AWS instance metadata and credentials, and three medium-severity vulnerabilities in Anthropic's Git MCP that, when chained with the Filesystem MCP and exploited via indirect prompt injection, can result in remote code execution; analysis of thousands of MCP servers suggests a large exposure surface (roughly 36.7% potentially affected).
Your team is not currently subscribed to this feed. You must subscribe to it in order to see this post.