Secure AI Atlas mark Secure AI Atlas SECURITY & GOVERNANCE

Framework

OWASP Top 10 for Agentic AI Applications 2025

A dedicated top-10 list addressing the unique security challenges of agentic AI systems — where LLMs plan, execute multi-step tasks, invoke tools, and operate with delegated authority over enterprise systems.

Agentic AIVulnerability ManagementApplication SecurityStandard

The OWASP Top 10 for Agentic Applications extends the LLM Top 10 to address the distinct threat surface of autonomous, tool-using, multi-step AI agents. While the LLM Top 10 focuses on language-model applications, the Agentic Top 10 targets systems that plan, remember, invoke APIs, and execute actions on behalf of users.

The ten agentic risks

#RiskCore concern
AA01Prompt InjectionExternal or internal untrusted content becomes instructions that the agent follows
AA02Excessive AgencyThe agent is granted authority beyond what the task, context, or system reliability can justify
AA03Insecure Tool InvocationTools receive malformed, malicious, or unvalidated input from an agent that trusts its own reasoning
AA04Agent-to-Agent Trust ExploitationMulti-agent systems inherit trust assumptions across agent boundaries without verification
AA05Memory PoisoningPersistent memory or context stores are contaminated with malicious data that influences future decisions
AA06Supply Chain CompromiseAgent dependencies — plugins, MCP servers, tool registries — introduce untrusted execution paths
AA07Goal MisalignmentAgent pursues a proxy or sub-goal that diverges from the intended objective, amplified by autonomy
AA08Output PropagationAgent outputs are consumed by downstream systems as trusted inputs without validation
AA09Credential and Session HijackingAgent-held credentials, tokens, or sessions are stolen and reused across systems and time
AA10Unbounded ExecutionAgent runs without time, cost, or scope limits, enabling runaway resource consumption or damage amplification

ATLAS connection

The Agentic Top 10 maps to four core ATLAS risks: Excessive Agency, Insecure Tool Invocation, Prompt Injection, and Agentic Supply Chain Compromise. ATLAS controls — including Human Approval for High-Impact Actions, Off-Agent Authorization for Tool Calls, and Prompt and Output Logging — provide the operational countermeasures.

Official website →

Official source: OWASP