A critical code injection vulnerability in QWED-MCP allows unauthenticated attackers to execute arbitrary OS commands through the math verification library.

The library suffers from improper control of code generation (CWE-94) because the verify_math_expression function passes unsanitized input to Python's eval function. An unauthenticated attacker can exploit this if they can influence input passed to this library function, leading to full process-level command execution.

With a CVSS score of 9.8, this vulnerability poses a catastrophic risk to confidentiality, integrity, and availability. An attacker can gain the permissions of the user running the qwed-mcp process, enabling them to steal secrets, modify critical system data, or pivot to internal network services.

Immediate Action: Update the QWED-AI qwed-mcp library to version 0.2.1 or later to ensure proper sanitization of mathematical expressions.

Proactive Monitoring: Audit downstream integrations that pass user-supplied input to the library and review system logs for suspicious process spawning or unauthorized file access.

Compensating Controls: Implement strict input validation at the application layer to block Python-specific syntax or built-in function calls before input reaches the QWED-MCP library.

Public Exploit Available: Unknown

The reliance on insecure evaluation functions makes this library highly susceptible to exploitation. Administrators should identify all applications utilizing the QWED-MCP library and apply the version 0.2.1 update without delay to mitigate the risk of arbitrary code injection.