Skip to main content
Mallory
← Back to News
ai-platform-securityopen-source-dependency-vulnerabilityunmanaged-asset-discovery

OX Security Flags RCE and Server Takeover Risks Across the MCP Ecosystem

Emerging Technology Risk · 4 sources · Updated 4d ago · First seen Nov 12, 2026

OX Security Flags RCE and Server Takeover Risks Across the MCP Ecosystem

OX Security’s analysis of more than 15,000 Model Context Protocol (MCP) servers in public registries and 3,000 open-source AI projects identified remote command execution risks, abandoned servers susceptible to takeover, and data sovereignty concerns. The findings concern infrastructure that connects AI agents and coding assistants to tools and data, potentially exposing organizations through the integrations those assistants use. The research was discussed in an SC Media podcast segment.

OX Security also describes an architectural flaw at MCP’s core as a systemic AI supply-chain vulnerability in its research article. The discussion emphasizes discovering, governing, and securing organizational MCP usage at scale. Security teams should inventory MCP integrations, assess server ownership and maintenance, restrict tool permissions, and review where connected services process sensitive data. The supplied references do not establish a confirmed breach or identify specific affected versions or fixes.

New surface, new adversariesTrack how attackers are adapting to this technology.
Track this risk
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

2 events · Nov 5, 2026
  1. Reported Nov 12, 20265d ago

    OX Security identifies security risks across the MCP ecosystem

    OX Security analyzed more than 15,000 MCP servers on public registries and 3,000 open-source AI projects. The analysis identified remote command execution risks, abandoned servers susceptible to takeover, and data sovereignty risks.

    Podcast Segment: Model Context Points of Failure: MCP Security Meets Scale - Moshe Siman Tov Bustan - OWSP26 #1 | SC Media
  2. Nov 5, 202613d ago

    OX Security details MCP command-execution risks and Anthropic's mitigation refusal

    At OWASP Global AppSec, OX Security researcher Moshe Siman Tov Bustan detailed how local MCP STDIO configurations could execute arbitrary commands with the logged-in user's permissions. He reported that Anthropic considered the behavior intentional and declined mitigation, while at least 10 associated downstream CVEs had been created and some affected software remained unpatched or unmitigated.

    OWASP Global AppSec: MCP security still a mind-boggling mess | resource | SC Media
Linked entities

Related entities

Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.

Open in app →

Organizations

2 linked
International Business MachinesAnthropic
Do you run this?Follow how adversaries are adapting to this technology, and where it touches your stack today.
Track This Risk
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.