SSRF in binary-husky/gpt_academic

The SSRF vulnerability in gpt_academic presents a significant risk because it allows attackers to bypass security controls and access internal resources. By exploiting the HotReload plugin, an attacker can craft malicious requests that the server will execute, effectively acting as the server itself. This can lead to the exposure of sensitive data, such as API keys, database credentials, or internal network configurations. Furthermore, an attacker could potentially use the compromised server as a launchpad for lateral movement within the network, gaining access to other systems and data. The lack of proper input validation on the URL allows for arbitrary hostnames to be specified, greatly expanding the potential attack surface.

Organizations deploying gptacademic in environments with sensitive internal resources are at significant risk. This includes deployments where the Gradio Web server has access to API keys, database credentials, or other confidential information. Shared hosting environments where multiple users share the same gptacademic instance are also particularly vulnerable, as an attacker could potentially exploit the vulnerability to access data belonging to other users.

• python / server:

import requests
from urllib.parse import urlparse

def check_ssrf(url):
    try:
        parsed_url = urlparse(url)
        if parsed_url.scheme in ('http', 'https'):
            # Check if the hostname is internal or unexpected
            if any(keyword in parsed_url.hostname for keyword in ['localhost', '127.0.0.1', '192.168.']):
                return True
    except Exception as e:
        print(f"Error parsing URL: {e}")
    return False

# Example usage
url_to_check = input("Enter URL to check: ")
if check_ssrf(url_to_check):
    print("Potential SSRF vulnerability detected!")
else:
    print("No SSRF vulnerability detected.")

• generic web:

curl -I 'http://<gpt_academic_server>/hotreload?url=http://internal-resource/'

Examine the response headers and body for signs of internal resource access or unexpected behavior.

1. 2025-03-20Disclosure

   disclosure

1. Gereserveerd2024-11-08
2. Gepubliceerd2025-03-20
3. Gewijzigd2025-07-15
4. EPSS bijgewerkt2026-04-02

The primary mitigation for CVE-2024-11031 is to upgrade to a patched version of gptacademic as soon as it becomes available. Until a patch is deployed, implement temporary workarounds to reduce the attack surface. These include configuring a Web Application Firewall (WAF) to block requests containing suspicious URLs or patterns. Network segmentation can also limit the potential impact by isolating the gptacademic server from sensitive internal resources. Implement strict URL validation within the HotReload plugin to ensure that only trusted domains are accessed. Consider disabling the HotReload plugin entirely if it is not essential for the application's functionality.