Browse / Improper Output Handling / CVE-2023-36095
langchain Code Injection vulnerability
- Class
- Improper Output Handling (filed by a model that read the advisory)
- Technique
- Code and command execution
- Also involves
- Tool and Agent Misuse
- Affected
- langchain (PyPI)
- CVSS
- CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
- EPSS
- 1.2% chance of exploitation in 30 days
How it works, in short
The PALChain math prompt passes model-generated code to Python exec, so whatever the model writes runs on the host. (Our summary, written by a model. The source text is below.)
Details
An issue in Harrison Chase langchain allows an attacker to execute arbitrary code via the PALChain,from_math_prompt(llm).run in the python exec method.
References
- https://nvd.nist.gov/vuln/detail/CVE-2023-36095
- https://github.com/langchain-ai/langchain/issues/5872
- https://github.com/langchain-ai/langchain/pull/6003
- https://github.com/langchain-ai/langchain/pull/7870
- https://github.com/langchain-ai/langchain/commit/8ba9835b925473655914f63822775679e03ea137
- https://github.com/langchain-ai/langchain/commit/e294ba475a355feb95003ed8f1a2b99942509a9e
- https://github.com/hwchase17/langchain
- https://github.com/langchain-ai/langchain
- https://github.com/langchain-ai/langchain/commits/v0.0.236?after=4d8b48bdb3f17c764c5c2e3c7140071603869e74+34&branch=v0.0.236&qualified_name=refs%2Ftags%2Fv0.0.236
- https://github.com/pypa/advisory-database/tree/main/vulns/langchain/PYSEC-2023-138.yaml
- http://langchain.com
Source: GitHub Advisory Database, used under CC BY 4.0.