Claude Certified Architect – Foundations Practice Question 74: Skill Arguments User Invoked
Claude Certified Architect, Foundations · Test 8 · question 4 of 10
A team wants a reusable Claude Code workflow for fixing a GitHub issue: read the issue, implement the fix, add tests and create a commit. Developers should start it explicitly by typing the issue number, as in "/fix-issue 123", and Claude must never start it on its own because it creates commits. Which setup meets this?
Show answer and explanation
Answer:
A skill is invoked as /skill-name, and whatever the user types after the name replaces the $ARGUMENTS placeholder in the skill content (so "/fix-issue 123" expands to "Fix GitHub issue 123 ..."). Setting disable-model-invocation: true means only the user can trigger it, which is the documented way to protect workflows with side effects such as commits.
- A is wrong: CLAUDE.md is always-loaded context, not an invocable command, and {{issue}} is not a substitution syntax Claude Code performs. - B is wrong: user-invocable: false is the opposite control; it hides the skill from the user's slash menu and leaves Claude able to invoke it automatically. - D is wrong: a hook is custom code that can block a prompt or add context, but it cannot create a proper command with argument substitution, and it does not stop Claude invoking anything.
Reference: Extend Claude with skills