CVE-2023-37274 Python code execution sandbox escape in non-docker version in Auto-GPT
Path traversal in Auto-GPT allows overwriting host files to escape sandbox and execute arbitrary code on the host system.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Local or same network only
- Victim action needed
- A user has to click or open something
- Fix
- None listed
What is CVE-2023-37274?
CVE-2023-37274 is a high-severity path traversal vulnerability in Significant-Gravitas Auto-GPT, published on 13 Jul 2023. It can only be exploited from the same machine or local network, and only if a user opens a file or follows a link.
Official description
Auto-GPT is an experimental open-source application showcasing the capabilities of the GPT-4 language model. When Auto-GPT is executed directly on the host system via the provided run.sh or run.bat files, custom Python code execution is sandboxed using a temporary dedicated docker container which should not have access to any files outside of the Auto-GPT workspace directory. Before v0.4.3, the `execute_python_code` command (introduced in v0.4.1) does not sanitize the `basename` arg before writing LLM-supplied code to a file with an LLM-supplied name. This allows for a path traversal attack that can overwrite any .py file outside the workspace directory by specifying a `basename` such as `../../../main.py`. This can further be abused to achieve arbitrary code execution on the host running Auto-GPT by e.g. overwriting autogpt/main.py which will be executed outside of the docker environment meant to sandbox custom python code execution the next time Auto-GPT is started. The issue has been patched in version 0.4.3. As a workaround, the risk introduced by this vulnerability can be remediated by running Auto-GPT in a virtual machine, or another environment in which damage to files or corruption of the program is not a critical problem.
Weakness: CWE-94
Which versions are affected by CVE-2023-37274?
It affects 1 product.
Products include Agpt Autogpt Classic.
How do I fix CVE-2023-37274?
No fix has been published for CVE-2023-37274 yet. Limit exposure of the affected software until one is available.
Upgrade, and resolve every user-supplied path against an allowed base folder before opening it. More about Path traversal.
Is CVE-2023-37274 being exploited?
There is no known exploitation of CVE-2023-37274 in the wild and no public exploit code.
Other vulnerabilities in Significant-Gravitas Auto-GPT
- CVE-2024-6091 Shell Command Denylist Bypass in significant-gravitas/autogpt
- CVE-2024-8156 Command Injection in significant-gravitas/autogpt
- CVE-2024-1879 CSRF to RCE in significant-gravitas/autogpt
- CVE-2024-1881 Improper Neutralization of Special Elements used in an OS Command in significant-gravitas/autogpt
- CVE-2023-37273 Docker escape in Auto-GPT when running from docker-compose.yml included in git repo
Sources
Also known as GHSA-5h38-mgp9-rj5f.