Checked fact 99510 Oct 2026Safety and security

Anthropic's documentation says: "Put untrusted content only in tool results." and says Claude is trained to treat instructions that appear inside tool results with appropriate skepticism.

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

Put untrusted content only in tool results. Deliver third-party content to Claude inside tool_result blocks, never in system prompts or plain user text blocks. Claude is trained to treat instructions that appear inside tool results with appropriate skepticism.

What the source said when we opened it, on 10 Oct 2026.

The source

Mitigate jailbreaks and prompt injections
Anthropic (Claude Platform Docs)

Open the source archived copy

Checked

Checked by the notis newsroom on 10 Oct 2026, against the source above.

In the story

How prompt injection works and how to limit the damage 10 Oct 2026

Cite this fact

Short permanent address

In a sentence

“Anthropic's documentation says: "Put untrusted content only in tool results." and says Claude is trained to treat instructions that appear inside tool results with appropriate skepticism.” ai notis, checked 10 Oct 2026. Source: Mitigate jailbreaks and prompt injections (Anthropic (Claude Platform Docs)). https://ainotis.com/c/995

As a reference (APA style)

ai notis. (2026, October 10). Anthropic's documentation says: "Put untrusted content only in tool results." and says Claude is trained to treat instructions that appear inside tool results with appropriate skepticism [Checked claim 995]. ai notis. https://ainotis.com/c/995

Anyone may quote this address. It does not change; if we correct the story, this page says so.