---
id: "willison-prompt-injection-design-patterns"
title: "Design Patterns for Securing LLM Agents against Prompt Injections"
creator: "Simon Willison"
creator_url: "https://simonwillison.net/2025/Jun/13/prompt-injection-design-patterns/"
url: "https://simonwillison.net/2025/Jun/13/prompt-injection-design-patterns/"
level: "advanced"
level_max: null
category: "security-and-red-teaming"
subcategory: "agent-and-tool-security"
region: "West"
license: "closed"
license_scope: "linked-resource"
models: ["any"]
source_date: "2025-06-13"
date_added: "2026-10-18"
---

[Security And Red Teaming](/security-and-red-teaming/index.md)

# Design Patterns for Securing LLM Agents against Prompt Injections

by [Simon Willison](https://simonwillison.net/2025/Jun/13/prompt-injection-design-patterns/)

[Read the original →](https://simonwillison.net/2025/Jun/13/prompt-injection-design-patterns/)

Summary of a paper's defensive agent architectures.

Willison reviews published work on architectural defenses against prompt injection. The entry credits his commentary rather than presenting it as the underlying paper, and includes no operational attack examples.

Level: [Advanced](/levels/index.md#advanced)
Models: any
Category: [Security And Red Teaming](/security-and-red-teaming/index.md)
Tags: [security](/tags/security/index.md), [prompt-injection](/tags/prompt-injection/index.md)
Source date: 2025-06-13
Added: 2026-10-18
Last verified: 2026-10-18
Status: ok
Prerequisites: Requires implementation experience or critical reading of technical methods and system trade-offs.
Subcategory: [agent and tool security](/security-and-red-teaming/agent-and-tool-security/index.md)
Region: West
Original language: en
Access: free
Resource reuse: closed: No general open-reuse grant established from the inspected page. Closed is the conservative index bucket, not a legal finding; the resource is publicly readable. [Facet definitions](/about/index.md#facets)

Summary in our own words. Rights remain with the original creator. [Credit and correction policy](/about/index.md#credit-policy).

[Plain-text Markdown copy](/security-and-red-teaming/agent-and-tool-security/willison-prompt-injection-design-patterns/index.md)