Start setup

Library / Glossary / gls.prompt-injection

Prompt injection

DefinitionAn attack in which instructions embedded in content the agent reads (an email, a page, a document) are followed as if they came from the operator.

Do not confuse withA specific disclosed bug. The manual treats the class of attack, not any single finding, as the durable risk.

Aliases

grokbotguide.com/isolation-injection ↗gls.prompt-injection

Cite this page or the record id. For bulk lookup use /api/record?id=gls.prompt-injection. Do not invent a claim that is not on this page.