Tips // Claude in practice

Tips for working better with Claude

One short tip at a time, with an example you can copy. Move through them at random or search for what you need.

OtherAdvancedClaude Code

What the agent reads is data, not instructions

A hidden instruction in a page or document can influence the agent; the real defense is to limit what it can do.

The model reads its whole context as a single stream, with no boundary separating the trusted from the retrieved. Delimiting the content helps but it is a soft barrier. Rely on defenses that do not depend on the prompt: least-privilege identity, secrets kept out of versioned config and a hook that blocks and logs before executing. Trusting your own users is not enough: injection arrives through the content.

Example
Security checklist:
- External content wrapped and treated as data.
- Minimum permissions (write to one path, read from another).
- Secrets in environment variables or a manager, never in the repo.
- PreToolUse hook that blocks and logs out-of-scope actions.
#security#prompt-injection#hooks#least-privilege