Paths/Claude Engineer/Keeping Claude Code Safe / Untrusted Content
Spot a Prompt Injection
An issue comment that says ignore your instructions is an attack, and the agent will read it as an instruction. Write the rules for untrusted text.
Validator
⏱ 0:00
IDLE — checks run live as you type commands; submit when done
01Requirements
live checks- File ~/.claude/security/untrusted.md exists○
- The document states which inputs are untrusted○
- The document states what the agent does with them○
02Walkthrough
1 of 4 revealed- 01
Create the security directory
The rules live with the permission set.
mkdir -p ~/.claude/security - 3 locked · each failed submit reveals one
03Stuck?
0 of 3 revealedTake a hint without losing the lab — each one points at the thing to look at, not at the answer.
04AI Tutor
guides, never solvesTell me what you're stuck on — I'll nudge, not solve. Which requirement are you looking at?
04Notebook
saved locally in this browserlearner@lab — ubuntu:24.04
Session is disposable
↑/↓ recalls commands · Ctrl+Enter submits · type help for the command list