An unauthenticated code injection vulnerability in MervinPraison PraisonAI versions 4.5.139 through 4.6.31 allows remote attackers to execute arbitrary code, creating a severe risk of complete system compromise.

This flaw is classified as a Code Injection vulnerability under CWE-94, occurring because an import sink in praisonai/templates/tool_override.py lacks proper access controls and is reached during recipe execution. The attacker can trigger this vulnerability unauthenticated via the POST /v1/recipes/run endpoint.

A successful exploit of this vulnerability grants the attacker full control over the host system, leading to potential data exfiltration, system downtime, and severe reputational damage. With a CVSS score of 8.4, the risk level is high, reflecting the potential for total loss of confidentiality, integrity, and availability.

Immediate Action: Update PraisonAI immediately to version 4.6.32 or later where the import sink is properly guarded.

Proactive Monitoring: Monitor API access logs for suspicious POST requests directed at the /v1/recipes/run endpoint, particularly those referencing unusual local paths or external GitHub repositories.

Compensating Controls: Deploy a Web Application Firewall rule to inspect and block unauthorized requests to the recipe execution endpoint until updates can be applied.

Public Exploit Available: No - no confirmed public exploit or weaponized module is present in the current threat intelligence feeds.

The severity and remote attack vector of this vulnerability demand immediate attention from IT administrators. Apply the available patch to version 4.6.32 immediately to eliminate the exposure and protect the environment from potential exploitation.