Langroid is a Python framework designed to simplify the development of multi-agent AI applications, with its vulnerability profile centered on the core product and characterized by code-injection risks inherent to dynamic code generation and XML processing. The recurring weakness classes—improper code generation control and unrestricted XML entity reference handling—reflect the framework's integration of user input with executable contexts typical of agent-oriented and data-interchange layers. Current severity, exploitation activity, and exposure counts are shown alongside this summary.
The number and severity of CVEs published that impact products developed by Langroid over time
Signals from CVEs in this vendor scope (4 CVEs).
4 CVEs · Highest risk first
| CVE | Published | CVSS | Risk | KEV | Exploit |
|---|---|---|---|---|---|
CVE-2026-25481CRITICAL 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 p | Feb 4, 2026 | 9.6 | 30 | NO | NO |
CVE-2025-46724CRITICAL Langroid is a Python framework to build large language model (LLM)-powered applications. Prior to version 0.53.15, `TableChatAgent` uses `pandas eval()`. If fed by untrusted user i | May 20, 2025 | 9.8 | 28 | NO | NO |
CVE-2025-46725CRITICAL Langroid is a Python framework to build large language model (LLM)-powered applications. Prior to version 0.53.15, `LanceDocChatAgent` uses pandas eval() through `compute_from_docs | May 20, 2025 | 9.8 | 26 | NO | NO |
CVE-2025-46726CRITICAL Langroid is a framework for building large-language-model-powered applications. Prior to version 0.53.4, a LLM application leveraging `XMLToolMessage` class may be exposed to untru | May 5, 2025 | 9.1 | 26 | NO | NO |
Signals from CVEs in this vendor scope (4 CVEs).
An overview of all social media posts that mention a CVE ID that affects a product developed by Langroid.
Media articles that mention a CVE ID that affects a product developed by Langroid — matched by CVE ID, not by vendor name.