PyPI: langroid

CVE-2026-55615

Safety vulnerability ID: SFTY-20260706-57600

Affected versions of the Langroid package are vulnerable to Command Injection due to the execution of LLM-generated Cypher queries without validation. The `Neo4jChatAgent` component passes user-influenced Cypher queries directly to the Neo4j driver through methods like `read_query` and `write_query`, without implementing any validation or statement-type restrictions. An attacker can exploit this vulnerability by manipulating the prompt to execute arbitrary Cypher commands, potentially leading to unauthorized data access, data destruction, and, when certain configurations are enabled, remote code execution.

Created at: Jul 10, 2026Updated at: Jul 10, 2026

Overview

Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-conditional RCE), mirroring the SQLChatAgent bug fixed in CVE-2026-25879

Advisory

langroid – Improper Neutralization of Special Elements in Output

Affected Package

Affecting langroid package, versions<= 0.65.4

Also affects

---

How to Fix

Upgradelangroidto0.65.5or higher.

Mitigation and Workarounds

---

Vulnerable Functions

Functions linked to known vulnerabilities.

Vulnerable function data is available for Enterprise customers

Book a call with us to see Safety in action.

Safety

Verified by Safety

Our Cybersecurity Intelligence Team reviewed this vulnerability. We combine public data with our own research to find issues not yet reported to public sources.

Learn more