How to resolve
Implement an emergency stop mechanism to immediately halt agent execution across all endpoints. Maintain a centralized policy to disable agents organization-wide.Risk
Governance/Compliance This vulnerability falls under ASI10:2026 — Rogue Agents in the OWASP Top 10 for Agentic Applications. Compromised or misaligned agents act outside their intended scope without external triggering. Security Agent system has no mechanism to immediately halt or terminate agent execution. If exploited, this can compromise the agent’s integrity, confidentiality, or availability, potentially affecting downstream systems and data.Explanation
Agent system has no mechanism to immediately halt or terminate agent execution. This falls under ASI10 (Rogue Agents): Compromised or misaligned agents act outside their intended scope without external triggering.Specifications
Trigger- Agent scan detects agent system has no mechanism to immediately halt or terminate agent execution
- no max_iter or timeout on agent
- no signal handler or shutdown mechanism
- no graceful termination path in agent code
- HIGH