# AI Agent Security: The OWASP Agentic Top 10 Guide

DevFeed: [AI Agent Security: The OWASP Agentic Top 10 Guide](<https://devfeed.tech/articles/ai-agent-security-the-owasp-agentic-top-10-guide-116611.md>)

Original publisher: [Read original article](<https://www.harness.io/blog/ai-agent-security-applying-the-owasp-top-10-across-the-software-development-lifecycle>)

Author: Renny Shen

Published: 2026-11-04T00:00:00Z

Content type: article

Language: en

Sources: [Harness](<https://devfeed.tech/sources/harness-blog.md>)

Topics: [LLM security](<https://devfeed.tech/topics/llm-security.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [AI software supply chain security](<https://devfeed.tech/topics/ai-software-supply-chain-security.md>), [DevSecOps](<https://devfeed.tech/topics/devsecops.md>), [Application Security](<https://devfeed.tech/topics/application-security.md>), [code security](<https://devfeed.tech/topics/code-security.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>)

Tags: [agent-communication](<https://devfeed.tech/tags/agent-communication.md>), [agent-identities](<https://devfeed.tech/tags/agent-identities.md>), [agent-security](<https://devfeed.tech/tags/agent-security.md>), [agent-skills](<https://devfeed.tech/tags/agent-skills.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent-security](<https://devfeed.tech/tags/ai-agent-security.md>), [ai-agent-security-risks](<https://devfeed.tech/tags/ai-agent-security-risks.md>), [ai-agent-threats](<https://devfeed.tech/tags/ai-agent-threats.md>), [ai-apis](<https://devfeed.tech/tags/ai-apis.md>), [ai-bill-of-materials](<https://devfeed.tech/tags/ai-bill-of-materials.md>), [ai-firewall](<https://devfeed.tech/tags/ai-firewall.md>), [ai-generated-code](<https://devfeed.tech/tags/ai-generated-code.md>), [api-keys](<https://devfeed.tech/tags/api-keys.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [development](<https://devfeed.tech/tags/development.md>), [owasp](<https://devfeed.tech/tags/owasp.md>), [owasp-top-10](<https://devfeed.tech/tags/owasp-top-10.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [testing](<https://devfeed.tech/tags/testing.md>)

## AI overview

AI agent security requires controls across the lifecycle: development and pipeline checks reduce preventable weaknesses, while runtime safeguards address threats that emerge as agents encounter content and take actions. The text recommends inventorying agents, tools, MCP servers, identities, credentials, and permissions; scanning dependencies and updates, pinning versions, detecting exposed secrets, and reviewing executable configurations before deployment; and monitoring tool use and behavior at runtime. Risks discussed include supply chain compromise, tool misuse, privilege abuse, code execution, goal hijacking, poisoned memory, insecure agent communication, cascading failures, rogue agents, and overtrust. Suggested runtime measures include least privilege, short-lived credentials, human confirmation for high-impact actions, isolated execution, policy checks, limits, rollback paths, and a clear kill switch. The article presents OWASP's Agentic Top 10 as a useful, non-exhaustive framework and describes Harness capabilities spanning scanning, testing, discovery, and runtime protection.

## Source excerpt

Learn how to secure AI agents across development, testing, deployment, and runtime using the OWASP Top 10 for Agentic Applications. | Blog