From base-tools
Template skill — copy this directory and edit. Replace this description with WHEN Claude should invoke the skill (e.g. "Use when the user asks to inspect X" or "Triggers on mentions of Y"). The description is the only signal Claude uses to auto-load the skill, so be specific about the trigger surface.
How this skill is triggered — by the user, by Claude, or both
Slash command
/base-tools:example-skillThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **This file is a template.** Duplicate the `example-skill/` directory under `skills/`, rename it, and replace every section below.
This file is a template. Duplicate the
example-skill/directory underskills/, rename it, and replace every section below.
One paragraph: what problem this skill solves, what domain knowledge it encodes, and what it expects from the user (inputs, preconditions, VPN, credentials, etc.).
Optional but valuable: list situations where Claude should NOT auto-invoke this. Helps Claude avoid mis-firing.
Write the body in the imperative voice — these are instructions Claude will follow when the skill is active.
Tables, endpoint lists, code snippets, command templates — anything Claude would need to look up while the skill is active.
| Column | Column |
|---|---|
| value | value |
If the skill enables destructive operations (writes, deletes, cluster ops), document the confirmation gates here — name the exact preconditions Claude must check, the dry-run requirement, and the user-confirmation echo before any irreversible action.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub blazemeter/claude-base --plugin base-tools