VYPR
Critical severity9.6NVD Advisory· Published Feb 4, 2026· Updated Jun 17, 2026

CVE-2026-25481

CVE-2026-25481

Description

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.59.32, there is a bypass to the fix for CVE-2025-46724. TableChatAgent can call pandas_eval tool to evaluate the expression. There is a WAF in langroid/utils/pandas_utils.py introduced to block code injection CVE-2025-46724. However it can be bypassed due to _literal_ok() returning False instead of raising UnsafeCommandError on invalid input, combined with unrestricted access to dangerous dunder attributes (__init__, __globals__, __builtins__). This allows chaining whitelisted DataFrame methods to leak the eval builtin and execute arbitrary code. This issue has been patched in version 0.59.32.

AI Insight

LLM-synthesized narrative grounded in this CVE's description and references.

Affected packages

Versions sourced from the GitHub Security Advisory.

PackageAffected versionsPatched versions
langroidPyPI
< 0.59.320.59.32

Affected products

3
  • cpe:2.3:a:langroid:langroid:*:*:*:*:*:*:*:*+ 1 more
    • cpe:2.3:a:langroid:langroid:*:*:*:*:*:*:*:*range: <0.59.32
    • (no CPE)range: < 0.59.32
  • ghsa-coords
    Range: < 0.59.32

Patches

Vulnerability mechanics

References

7

News mentions

0

No linked articles in our index yet.