A Skill Can Run Commands Before Claude Reads It: What to Check First
You find a skill on GitHub that writes release notes. It looks harmless: a SKILL.md with a description and a few lines of instructions. You copy the folder into .claude/skills/ , ask for release notes, and it works. What you may not have checked is that a skill isn't only text. In Claude Code, a few lines in that file can run shell commands before Claude reads anything, pre-approve tools, or…
You have discovered a GitHub skill that writes release notes and copied it into your Claude Code skills directory. Upon running it, you were surprised to learn that the skill folder contains more than just plain text. In Claude Code, certain elements within the folder can execute shell commands before Claude reads the skill's content, pre-approve tools, or register hooks that remain active throughout the session.
This behavior is not immediately evident from the instructions, but it is crucial to understand before installing someone else's skill.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
