CVE-2025-6514
OS command injection in mcp-remote when connecting to untrusted MCP servers
Description
mcp-remote is exposed to OS command injection when connecting to untrusted MCP servers due to crafted input from the authorization_endpoint response URL
INFO
Published Date :
July 9, 2025, 1:15 p.m.
Last Modified :
April 15, 2026, 12:35 a.m.
Remotely Exploit :
Yes !
Source :
[email protected]
Affected Products
The following products are affected by CVE-2025-6514
vulnerability.
Even if cvefeed.io is aware of the exact versions of the
products
that
are
affected, the information is not represented in the table below.
No affected product recoded yet
CVSS Scores
| Score | Version | Severity | Vector | Exploitability Score | Impact Score | Source |
|---|---|---|---|---|---|---|
| CVSS 3.1 | CRITICAL | [email protected] |
Solution
- Sanitize all inputs from authorization_endpoint.
- Validate the URL for the authorization endpoint.
- Update the mcp-remote library to the latest version.
Public PoC/Exploit Available at Github
CVE-2025-6514 has a 103 public
PoC/Exploit available at Github.
Go to the Public Exploits tab to see the list.
References to Advisories, Solutions, and Tools
Here, you will find a curated list of external links that provide in-depth
information, practical solutions, and valuable tools related to
CVE-2025-6514.
CWE - Common Weakness Enumeration
While CVE identifies
specific instances of vulnerabilities, CWE categorizes the common flaws or
weaknesses that can lead to vulnerabilities. CVE-2025-6514 is
associated with the following CWEs:
Common Attack Pattern Enumeration and Classification (CAPEC)
Common Attack Pattern Enumeration and Classification
(CAPEC)
stores attack patterns, which are descriptions of the common attributes and
approaches employed by adversaries to exploit the CVE-2025-6514
weaknesses.
We scan GitHub repositories to detect new proof-of-concept exploits. Following list is a collection of public exploits and proof-of-concepts, which have been published on GitHub (sorted by the most recently updated).
None
Python Standard ML Jinja
Security scanner for MCP/AI agent configurations
Python
Security auditor for Claude Code plugins, MCP servers, and agents — 23 rules, OWASP mapped, 100% offline
JavaScript YARA Shell
Score any MCP server before your team integrates it. No SDK installation required. Just a URL.
JavaScript TypeScript CSS
None
Dockerfile Python HTML
Atomadic Nexus SDK — Trust Layer for the Agent Economy. x402, lineage, hallucination, reputation.
Python
None
Trust Layer for the Agent Economy. x402, Stripe, lineage attestation, hallucination oracle, agent reputation. Codex-anchored confidence.
Python
Local-first firewall and black-box run reporter for MCP tool calls: block poisoned tools, detect drift, rate-limit agents, and audit runs.
agent-security ai-agents audit-logging developer-tools mcp model-context-protocol prompt-injection security-tools tool-poisoning typescript agent-observability local-first run-reports agent-tooling ai-security llm-security mcp-gateway mcp-security
JavaScript HTML TypeScript
adversarial security testing for MCP servers and agentic tool surfaces. built in rust.
Rust
Awesome Agentic MCP Security List
awesome awesome-list awesome-lists awesome-resources
MCP-BOM: a reproducible attack-surface benchmark for Model Context Protocol servers — NeurIPS 2026 ED submission
ai-security attack-surface benchmark bom mcp model-context-protocol
Python Makefile Dockerfile HTML
Paste your AGENTS.md, Cursor rules, Codex settings, MCP config, or other AI coding-agent configuration and get a plain-English security report in seconds. AgentGuard checks for 134 risky misconfiguration patterns, including the lethal trifecta, hardcoded secrets, unsafe shell and network access, webhook exfiltration, prompt leaks, cloud metadata SS
TypeScript CSS JavaScript
purplegate — Red/blue-team CI gate for agentic-AI apps. Prompt-injection, secrets, SAST, deps, IaC/RLS, workflow injection, MCP checks. One signed GitHub Action, mapped to OWASP LLM Top 10 v2025 + MITRE ATLAS.
agentic-ai ai-security ci-cd devsecops github-action llm-security mitre-atlas owasp owasp-llm-top-10 prompt-injection prompt-injection-detection sa sarif security supply-chain-security
Dockerfile Python
Security gateway for MCP traffic. 5 built-in rules block prompt injection, credential leakage, path traversal, cross-repo exfiltration, and destructive commands. One pip install, zero config.
mcp security python proxy claude ai-agents prompt-injection self-hosted model-context-protocol llm-security ai-security cursor security-tools path-traversal credential-leakage
Python
Results are limited to the first 15 repositories due to potential performance issues.
The following list is the news that have been mention
CVE-2025-6514 vulnerability anywhere in the article.
-
CybersecurityNews
MEDUSA Security Testing Tool With 74 Scanners and 180+ AI Agent Security Rules
MEDUSA, an AI-first Static Application Security Testing (SAST) tool boasting 74 specialized scanners and over 180 AI agent security rules. This open-source CLI scanner targets modern development chall ... Read more
-
The Hacker News
[Webinar] Securing Agentic AI: From MCPs and Tool Access to Shadow API Key Sprawl
Jan 13, 2026The Hacker NewsArtificial Intelligence / Automation Security AI agents are no longer just writing code. They are executing it. Tools like Copilot, Claude Code, and Codex can now build, t ... Read more
-
CybersecurityNews
Cloudflare Launches MCP Server Portals – A Unified Gateway to All MCP Servers
Cloudflare today launched MCP Server Portals in open beta, a groundbreaking capability designed to centralize, secure, and observe all Model Context Protocol (MCP) connections in an organization. By r ... Read more
-
The Cloudflare Blog
Securing the AI Revolution: Introducing Cloudflare MCP Server Portals
Securing the AI Revolution: Introducing Cloudflare MCP Server Portals2025-08-266 min read Securing the AI Revolution: Introducing Cloudflare MCP Server Portals Large Language Models (LLMs) are rapidly ... Read more
-
The Hacker News
Critical mcp-remote Vulnerability Enables Remote Code Execution, Impacting 437,000+ Downloads
Cybersecurity researchers have discovered a critical vulnerability in the open-source mcp-remote project that could result in the execution of arbitrary operating system (OS) commands. The vulnerabili ... Read more
-
CybersecurityNews
Critical mcp-remote Vulnerability Exposes LLM Clients to Remote Code Execution Attacks
A critical vulnerability CVE-2025-6514 with a CVSS score of 9.6 affecting the mcp-remote project allows attackers to achieve arbitrary operating system command execution on machines running mcp-remote ... Read more
The following table lists the changes that have been made to the
CVE-2025-6514 vulnerability over time.
Vulnerability history details can be useful for understanding the evolution of a vulnerability, and for identifying the most recent changes that may impact the vulnerability's severity, exploitability, or other characteristics.
-
New CVE Received by [email protected]
Jul. 09, 2025
Action Type Old Value New Value Added Description mcp-remote is exposed to OS command injection when connecting to untrusted MCP servers due to crafted input from the authorization_endpoint response URL Added CVSS V3.1 AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H Added CWE CWE-78 Added Reference https://github.com/geelen/mcp-remote/commit/607b226a356cb61a239ffaba2fb3db1c9dea4bac Added Reference https://jfrog.com/blog/2025-6514-critical-mcp-remote-rce-vulnerability Added Reference https://research.jfrog.com/vulnerabilities/mcp-remote-command-injection-rce-jfsa-2025-001290844/