logengineering notes by Delaa.
BlogProjectsAbout
All topics
  • all posts
  • symfony5
  • php5
  • security4
  • zf14
  • migration4
  • doctrine2
  • multi-tenancy2
  • architecture2
  • prompt-injection2
  • debugging1
  • background-jobs1
  • git1
  • workflow1
  • code-review1
  • http1
  • query1
  • rfc-100081
  • nextjs1
  • api-design1
  • github-actions1
  • ci-cd1
  • supply-chain1
  • llm1
  • ai-agents1
  • owasp1
  • legacy1
  • decision1
  • strangler-fig1
  • database1
  • postgresql1
  • authentication1
  • session1

Owasp

1 article

see also:securityprompt-injectionllm
  • 6a01426·3 Aug 2026·10 min read·in security·series: Indirect prompt injection

    Anatomy of Indirect Prompt Injection Hidden in Plain Text

    Twenty-two techniques, three delivery methods, and one structural weakness: AI assistants cannot distinguish instructions from data. Here is how attackers exploit that, and what works as a defense.

    securityprompt-injectionllmai-agentsowasp
log · Delaa · © 2026
RSSCategoriesSeriesTagsArchiveGitHubAbout