Instructlab: instructlab: arbitrary code execution due to hardcoded `trust_remote_code=true`
A flaw was found in InstructLab. The `linux_train.py` script hardcodes `trust_remote_code=True` when loading models from HuggingFace. This allows a remote attacker to achieve arbitrary Python code execution by convincing a user to run `ilab train/download/generate` with a specially crafted malicious model from the HuggingFace Hub. This vulnerability can lead to complete system compromise.
To mitigate this issue, only use models from trusted sources when performing `instructlab` operations. Review the origin and integrity of any HuggingFace model before using it with `ilab train/download/generate`. Consider running `instructlab` commands within a sandboxed or isolated environment to limit the potential impact of executing untrusted code.
Exploits
Credits
Red Hat would like to thank Martin Brodeur (independent security researcher) for reporting this issue.