← Concept Index

Agent & LLM security

Indirect prompt injection

Also called: poisoned content, injection via documents

DEFINITION

Injection that arrives not from the user but from content the agent fetches on its own — a booby-trapped web page, a shared doc, a calendar invite — often with the malicious text hidden from human eyes.

WHY IT MATTERS

This is the version that scales: an attacker never talks to you, they just plant instructions where your agent will read them. It's the reason 'the agent browsed a page' is a security event, not a convenience.

COMMONLY CONFUSED WITH

Direct prompt injection, where the user themselves types the malicious instruction. Here the user is the victim, not the source.

SOURCES

First-pass citations, limited to primary sources; a reviewer will broaden and verify these before this entry leaves draft.