---
title: "Discussion on \"Depth vs breadth: the two kinds of AI agent security testing\" | Hashnode"
description: "Discussion on \"Depth vs breadth: the two kinds of AI agent security testing\". Most agent security tooling focuses on finding new vulnerabilities. But who's checking whether last month's prompt injection still works after a model"
url: "https://hashnode.com/posts/depth-vs-breadth-the-two-kinds-of-ai-agent-security-testing/69b2165c6c896b0519d691a0"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Depth vs breadth: the two kinds of AI agent security testing](https://olegm.hashnode.dev/depth-vs-breadth-the-two-kinds-of-ai-agent-security-testing?utm_source=hashnode&utm_medium=feed)

Most agent security tooling focuses on finding new vulnerabilities. But who's checking whether last month's prompt injection still works after a model

Mar 12 · 7 min read · olegm.hashnode.dev  
Read the article

-   [#ai-agents](https://hashnode.com/tag/ai-agents)
-   [#security](https://hashnode.com/tag/security)
-   [#mcp](https://hashnode.com/tag/mcp)
-   [#ai-security](https://hashnode.com/tag/ai-security)
-   [#testing](https://hashnode.com/tag/testing)

## Written by

[Oleg Mukhin](https://hashnode.com/@olegm)

@olegm

  
[View profile](https://hashnode.com/@olegm)

## Responses

Markdown supported · ⌘↵ to post

No responses yet.