GHSA-22c2-9gwg-mj59
Langroid has a Code Injection vulnerability in LanceDocChatAgent through vector_store
Packages
Details
As written in the advisory (Markdown source, shown as plain text).
### Summary
[LanceDocChatAgent](https://github.com/langroid/langroid/blob/main/langroid/agent/special/lance_doc_chat_agent.py#L158) uses pandas eval() through `compute_from_docs()`:
https://github.com/langroid/langroid/blob/18667ec7e971efc242505196f6518eb19a0abc1c/langroid/vector_store/base.py#L136-L150
As a result, an attacker may be able to make the agent run malicious commands through [QueryPlan.dataframe_calc](https://github.com/langroid/langroid/blob/main/langroid/agent/special/lance_tools.py#L16) compromising the host system.
### Fix
Langroid 0.53.15 sanitizes input to the affected function by default to tackle the most common attack vectors, and added several warnings about the risky behavior in the project documentation.
References
External links from the advisory. They lead to third-party sites.