Slide 3 of 28
Part 1 — What Is It?Slide 3
Slide 3 · The Definition Part 1
What OWASP says — in plain English
The first half of the official definition, unpacked one phrase at a time.
📄 OWASP Top 10 for Agentic Applications 2026 · ASI02
Official definition — first half

"Agents can misuse legitimate tools due to prompt injection, misalignment, or unsafe delegation or ambiguous instruction — leading to data exfiltration, tool output manipulation or workflow hijacking."

What "prompt injection" means here

An attacker embeds instructions in content the agent is reading — a document, a web page, an email, an API response. The agent treats those instructions as legitimate commands and uses its tools accordingly. In the refund example on slide 1, the hidden instruction was a prompt injection.

What "misalignment" means here

The agent does something unintended not because of an attacker, but because its instructions were ambiguous or incomplete. A coding agent told to "clean up the repository" might delete files the developer still needed. No attack. Just a tool invoked beyond its intended scope.

What "unsafe delegation" means here

When one agent delegates a task to another, it may pass along its full tool permissions — even if the sub-agent only needs a small subset. A high-privilege orchestrator delegates to a narrow worker agent, which then has more capability than it should.

← Back Makes sense → What does the second half of the definition say?