standardDecember 2025ReviewedOpen access

OWASP Top 10 for Agentic Applications 2026

OWASP GenAI Security Project

OWASP GenAI Security Project

Abstract

Identifies the top 10 security risks specific to agentic AI applications including excessive agency, unsafe tool execution, and inadequate oversight.

Categories

Threat Modeling Risk Frameworks Agent Architecture

#OWASP#agentic#top-10#standard

Framework mappings

OWASP Top 10 for Agentic Applications

ASI01Agent Goal Hijack
ASI02Tool Misuse & Exploitation
ASI03Agent Identity & Privilege Abuse
ASI04Agentic Supply Chain Compromise
ASI05Unexpected Code Execution
ASI06Memory & Context Poisoning
ASI07Insecure Inter-Agent Communication
ASI08Cascading Failures
ASI09Human-Agent Trust Exploitation
ASI10Rogue Agents

Cite

@techreport{owasp2025owasp,
  title = {{OWASP Top 10 for Agentic Applications 2026}},
  author = {{OWASP GenAI Security Project}},
  year = {2025},
  month = dec,
  institution = {OWASP GenAI Security Project},
  url = {https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/}
}