CVE-2023-36188: LangChain remote code execution in PALChain Python exec method

Executive brief

LangChain, a popular framework for building applications with large language models, is vulnerable to a critical security flaw. An attacker can remotely execute malicious code on the system running the software by exploiting how it handles certain Python execution parameters. This could lead to a full system takeover, data theft, or disruption of services.

Technical details

A remote code execution (RCE) vulnerability exists in LangChain due to improper neutralization of special elements used in downstream components (CWE-74). The flaw is located within the PALChain parameter of the Python exec method, where externally-influenced input is insufficiently sanitized before being executed. A remote, unauthenticated attacker can exploit this by providing crafted input that triggers arbitrary command execution on the host system. The vulnerability is addressed in version 0.0.247.

Affected products

• langchain-ai langchain < 0.0.247

Timeline

• 2023-07-06: advisory
• 2023-07-06: disclosed
• 2023-07-06: patched

References

• https://api.github.com/users/cookesan
• https://github.com/cookesan
• https://api.github.com/users/cookesan/gists%7B/gist_id%7D
• https://api.github.com/users/cookesan/repos
• https://avatars.githubusercontent.com/u/6601329?v=4
• https://api.github.com/users/cookesan/events%7B/privacy%7D