CVE-2023-43654
CRITICAL EXPLOITED NUCLEITorchServe 0.1.0-0.8.1 - Server-Side Request Forgery via Model URL Parameter
Title source: llmExploitation Summary
CVE-2023-43654 has been observed exploited in the wild (reported by VulnCheck KEV).
EIP tracks 3 public exploits from researchers including OligoCyberSecurity, Idan Levcovich, Guy Kaplan, Gal Elbaz, Swapneil Kumar Dash, Spencer McIntyre, including a Metasploit module exploits/multi/http/torchserver_cve_2023_43654.
A Nuclei detection template is also available.
AI-analyzed exploit summary The repository contains a shell script named 'ShellTorchChecker.sh' but lacks technical details about CVE-2023-43654. The README is not provided in the snippet, but the presence of a generic script name and no exploit code suggests it may be a lure.
Description
TorchServe is a tool for serving and scaling PyTorch models in production. TorchServe default configuration lacks proper input validation, enabling third parties to invoke remote HTTP download requests and write files to the disk. This issue could be taken advantage of to compromise the integrity of the system and sensitive data. This issue is present in versions 0.1.0 to 0.8.1. A user is able to load the model of their choice from any URL that they would like to use. The user of TorchServe is responsible for configuring both the allowed_urls and specifying the model URL to be used. A pull request to warn the user when the default value for allowed_urls is used has been merged in PR #2534. TorchServe release 0.8.2 includes this change. Users are advised to upgrade. There are no known workarounds for this issue.
Exploits (3)
The repository contains a shell script named 'ShellTorchChecker.sh' but lacks technical details about CVE-2023-43654. The README is not provided in the snippet, but the presence of a generic script name and no exploit code suggests it may be a lure.
This repository provides a scanner tool (ShellTorch-Checker) to detect CVE-2023-43654, an SSRF vulnerability in PyTorch TorchServe that can lead to arbitrary remote code execution. It includes instructions for setting up a vulnerable environment and running the scanner to verify the presence of the vulnerability.
This Metasploit module exploits CVE-2022-1471 (SnakeYAML deserialization) and CVE-2023-43654 (PyTorch model server SSRF) to achieve remote code execution by registering a malicious MAR file containing a crafted YAML payload.
Nuclei Templates (1)
body="Requested method is not allowed, please refer to API document"
References (4)
Scores
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H