AppOmni launches AgentGuard to secure ServiceNow AI agents

AppOmni has introduced a real-time security solution for ServiceNow aimed at protecting organisations from threats arising from AI agents within the platform.

ServiceNow, used by 8,400 businesses globally, including a significant portion of the Fortune 500, is increasingly integrating AI to automate and assist in various business processes. The new security features from AppOmni address risks linked to AI agent behaviour, particularly those capable of collaborating and taking autonomous actions.

AI agent risks

Recent research from AppOmni has highlighted a vulnerability that allows attackers to exploit ServiceNow AI agents, even with some default security settings enabled. The risk involves attackers triggering second-order prompt injection attacks, allowing them to manipulate AI agents into stealing data, altering internal records, or providing broader access to systems. 

"This discovery is alarming because it isn't a bug in the AI, it's expected behavior as defined by certain default configuration options," said Aaron Costello, Chief of SaaS Security Research, AppOmni. "When agents can discover and recruit each other, a harmless request can quietly turn into an attack, with criminals stealing sensitive data or gaining more access to internal company systems. These settings are easy to overlook. If organizations using Now Assists AI agents aren't closely examining their configurations, they're likely already at risk."

AgentGuard launch

The new suite, AgentGuard, offers several capabilities focused on monitoring and securing AI agent activity in ServiceNow's Now Assist environment. It actively prevents prompt-injection attacks, flags and blocks incidents related to data loss prevention, and can quarantine users identified as malicious.

The tool extends to both native ServiceNow AI agents and custom-developed agents, providing security teams with real-time oversight and the ability to apply automated controls without interrupting legitimate workflows.

Visibility features

AgentGuard includes an agent inventory tool, providing organisations with detailed insight into which AI agents operate in their ServiceNow environment, their access permissions, and their behavioural patterns. This visibility is intended to help prevent over-permissioning and enforce the least-privilege principle.

Additionally, an automatic policy framework checks AI agent configurations for compliance with secure standards, providing alerts and guidance when risky or non-compliant settings are detected. Behaviour monitoring and alerting functions equip security teams to respond rapidly to suspicious actions.

Industry perspectives

The use of AI agents in enterprise platforms like ServiceNow is increasing, expanding the potential attack surface for cybercriminals. Organisations face the challenge of ensuring that autonomous agents, often designed to be highly collaborative, do not inadvertently expose sensitive data or allow unauthorised access.

"The proliferation of AI agents within critical business platforms like ServiceNow introduces a complex new attack surface that organizations are just beginning to grapple with," said Ying Ting Neoh, Industry Analyst at Frost and Sullivan. "AppOmni's AgentGuard and comprehensive AI agent security capabilities are vital for any enterprise looking to harness ServiceNow's AI power safely. By providing real-time threat detection and response, granular visibility, and automated posture enforcement, AppOmni empowers organizations to secure their Now Assist AI agent deployments, prevent sophisticated attacks like prompt injection, and maintain compliance without stifling innovation."