About this skill

Problem

When AI agents use external skills, teams often need to find the right skill, compare local and remote versions, install it into the agent workspace, update it, and publish changes. Skills on SkillHub may be scattered, local directories may contain stale versions, and the SkillHub CLI has both Python and npm variants, with login and publish limited to the Python version.

How It Works

This skill turns SkillHub operations into agent-triggerable workflows:

Discovery and installation: search, popular, and recommend locate skills, while install checks for the CLI and installs the selected skill.
Local maintenance: list shows installed skills, check-update compares versions, and update or update-all applies updates.
CLI and identity: cli-status reports CLI type, path, and version, while login, whoami, and logout manage credentials.
Publishing checks: publish --dry-run validates SKILL.md metadata; publishing requires a matching slug, an incremented version, and ownership.

Boundaries

Network access to the SkillHub API is required. Installing or updating skills can overwrite local files. The npm CLI is mainly for installation, while login and publishing depend on the Python CLI. Do not put Token values in SKILL.md or code, and confirm that the target slug belongs to the author before publishing.

Screenshots

Use Cases

Find a wallpaper-switching skill in an agent chat and auto-check the CLI before installing it locally.
Review installed skills, compare same-name remote versions, and batch-update stale local skills.
Prepare a local skill package for publishing by logging in, running a dry-run, and checking slug/version fields.
Diagnose CLI type, path, and login identity while handling npm versus Python CLI differences.

Best For

Engineers maintaining an agent skills directory: need a stable flow for search, install, update, and publishing.
Agent developers using SkillHub skill packs: want dialogue-triggered discovery, login, and publish dry runs.
Ops staff managing local agent skill environments: need CLI status, login identity, and npm/Python differences.
Authors preparing their skill packs: need dry-run checks for slug, version, and displayName before release.