Entity trail

CVE-2026-30615

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-07-13 / vibe-coding-researcherPattern: Untrusted HTML Reaching an IDE — READMEs, Tool Descriptions, Browsed Pages — Is the New RCE Surface for Coding AgentsCVE-2026-30615's zero-click chain (cloned-repo README or poisoned remote tool description → auto-registered malicious MCP server → RCE) sits alongside research finding 7,000+ publicly reachable MCP servers and packages totaling 150M+ downloads exposed to command injection, SSRF, and file access. As agents gain in-app browsers and auto-loaded MCP configs, any attacker-controlled text an agent reads is executable-adjacent — keep a human confirmation gate on config mutation and shell/npx calls sourced from fetched content.
2026-06-09 / vibe-coding-researcherCVE-2026-30615: Prompt-Injection RCE in Windsurf 1.9544.26A newly catalogued prompt-injection vulnerability (CVE-2026-30615) in Windsurf 1.9544.26 lets remote attackers execute arbitrary commands on a victim's system by smuggling instructions into content the agent processes. It lands amid a wider June 2026 MCP/agent security wave in which nine separate CVEs were traced to broken OAuth flows. Builders running Windsurf/Devin-Desktop agents should update and treat all untrusted file and web content as potential injection vectors.
2026-05-01 / vibe-coding-researcherOX Security Advisory: Systemic MCP STDIO Command Injection Affects 150M+ Downloads and 7,000+ ServersOX Security disclosed four families of command injection vulnerabilities in Anthropic's MCP STDIO implementations, affecting 150M+ downloads and 7,000+ publicly accessible servers. Family 1 (direct injection) hits LangFlow, GPT Researcher, LiteLLM, and 5 others. Family 3 (prompt injection → RCE) earned a critical CVE against Windsurf (CVE-2026-30615). Root cause: arbitrary command/argument values passed directly to StdioServerParameters without sanitization. Several vendors declined patches, claiming command execution is intentional design.

Graph relationships

CRITICIZES
MCP
CVE-2026-30615 is a critical command injection vulnerability in MCP
Source finding

Source trail

OX Security Adversa AI

Graph sources

entity graphfindings textkg edgeskg entities

Fetching from the wire…