Skip to main content
Agent-generated code or patches applied directly without human preview or sandboxed testing

​
How to resolve

Review all agent-generated code before execution. Run generated code in a sandboxed environment with restricted filesystem and network access.

​
Risk

Governance/Compliance This vulnerability falls under ASI05:2026 — Unexpected Code Execution in the OWASP Top 10 for Agentic Applications. Agent generates or runs code, commands, or scripts without safe sandboxing or review. Security Agent-generated code or patches applied directly without human preview or sandboxed testing. If exploited, this can compromise the agent’s integrity, confidentiality, or availability, potentially affecting downstream systems and data.

​
Explanation

Agent-generated code or patches applied directly without human preview or sandboxed testing. This falls under ASI05 (Unexpected Code Execution): Agent generates or runs code, commands, or scripts without safe sandboxing or review.

​
Specifications

Trigger
  • Agent scan detects agent-generated code or patches applied directly without human preview or sandboxed testing
The scanner looks for patterns such as:
  • code generation tool with auto-apply flag
  • generated patches run without review step
  • no human_input=True before code execution task
Severity
  • HIGH
Applies to: Agent