> ## Documentation Index
> Fetch the complete documentation index at: https://rootea.es/llms.txt
> Use this file to discover all available pages before exploring further.

# Exploiting AI agents to perform destructive actions

> PortSwigger Web Security Academy lab: Exploiting AI agents to perform destructive actions. Topic: LLM Attacks.

<p className="machine-summary"><span className="prompt"><code>\$ tldr</code></span> PortSwigger · LLM Attacks</p>

<div className="machine-meta">
  | · | · |
  | - | - |
  | Platform | PortSwigger Web Security Academy |
  | Topic | LLM Attacks |
  | Difficulty | <span className="dbadge dbadge-easy">APPRENTICE</span> |
  | Official lab | [Open](https://portswigger.net/web-security/llm-attacks/ai-powered-scanner-vulnerabilities/lab-indirect-prompt-injection-via-ai-powered-scan) |
</div>

## Solve the lab

| Language | Author | Format | Link |
| - | - | - | - |
| 🇬🇧 EN | **PortSwigger** | Texto | [Open](https://portswigger.net/web-security/llm-attacks/ai-powered-scanner-vulnerabilities/lab-indirect-prompt-injection-via-ai-powered-scan) |

## Resources by skill

| Skill | Source | Link |
| - | - | - |
| LLM Attacks (Prompt Injection) | PortSwigger | [Open](https://portswigger.net/web-security/llm-attacks) |

## Related skills

[LLM Attacks (Prompt Injection)](/en/skills/llm-attacks)

***

## Comments & tips

Solved this lab a different way? Share it here — comments live in [GitHub Discussions](https://github.com/FFuson/HTB_Writeups/discussions).

<div className="rootea-giscus-wrap" data-giscus-term="lab:llm-attacks-ai-powered-scanner-vulnerabilities-indirect-prompt-injection-via-ai-powered-scan" data-giscus-lang="en" />

*Last updated: 2026-05-08*

<script type="application/ld+json">
  {`{"@context":"https://schema.org","@type":"TechArticle","name":"Exploiting AI agents to perform destructive actions","headline":"Exploiting AI agents to perform destructive actions — PortSwigger lab index","url":"https://rootea.es/en/portswigger/labs/llm-attacks/ai-powered-scanner-vulnerabilities-indirect-prompt-injection-via-ai-powered-scan","inLanguage":"en","about":[{"@type":"Thing","name":"PortSwigger Web Security Academy"},{"@type":"Thing","name":"LLM Attacks"}],"isPartOf":{"@type":"WebSite","name":"rootea.es","url":"https://rootea.es"},"author":{"@type":"Organization","name":"rootea.es"}}`}
</script>