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.

SkillsIntermediateClaude.ai

Treat a Skill like software you install

Before enabling a third-party Skill, check who published it, what it could access and whether its scope is proportionate.

A Skill does not ask for permissions: it inherits the access of the session where it runs. Audit its contents (instructions, dependencies, files). Those from Anthropic or approved by your organization are the safest starting point, and an 'internal' one from another team is not automatically trustworthy either. Possible outcomes: enable, escalate to the admin or reject.

Example
Checklist: who publishes it? what would it reach in my sessions? does it do more than the task needs? If in doubt, escalate to your admin.
#skills#security#governance