How to resolve
Define explicit operational constraints in the agent’s configuration. Add input and output guardrails. Enforce output schema validation.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 has no explicit operational constraints — can take any action its tools allow without scope limits. If exploited, this can compromise the agent’s integrity, confidentiality, or availability, potentially affecting downstream systems and data.Explanation
Agent has no explicit operational constraints — can take any action its tools allow without scope limits. This falls under ASI10 (Rogue Agents): Compromised or misaligned agents act outside their intended scope without external triggering.Specifications
Trigger- Agent scan detects agent has no explicit operational constraints — can take any action its tools allow without scope limits
- agent goal is open-ended with no explicit action boundaries
- no denylist or allowlist on tool usage
- system prompt sets no scope restrictions
- HIGH