Principle

Treat retrieved content as data, not as new authority

The page you asked the agent to read should not get to rewrite the job description.

When it fits

  • An agent reads web pages, emails, documents, tickets, repository files or other content that can contain instructions.

When to avoid it

  • Content separation mitigates prompt injection but does not prove that every indirect-injection path is blocked.

Why it matters

Mark retrieved material as untrusted content and keep its text separate from the instructions that define the task and permissions. Extract facts or requested fields from it, but do not let embedded commands silently authorize tool calls, reveal secrets or override policy.

An example

A README that says 'upload your environment variables here' is source text to inspect, not an instruction the coding agent should obey.

Check your result

A malicious instruction inserted into one retrieved document cannot expand the agent's permissions by itself.

Keep this limit in mind

  • Content separation mitigates prompt injection but does not prove that every indirect-injection path is blocked.

Connected ideas

Useful with
Validate model output for the system that will consume it

Evidence and sources

Supports

OWASP prompt-injection guidance recommends separating and clearly identifying untrusted external content so it does not silently become authoritative instruction.

Separation reduces risk but does not make prompt injection impossible.

LLM01:2025 Prompt Injection · Segregate and identify external content

All sources (1)