AugmentClaude

Skills Installer

Install, list, validate, and package LiveAgent skills from various sources.

Installation

  1. Make sure Claude is on your device and in your terminal.

    Skills load from ~/.claude/skills/ when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then run claude in any terminal to verify.

    One-time setup
    npm i -g @anthropic-ai/claude-code

    Already have it? Skip ahead.

  2. Paste into Claude Code or into your terminal.

    This copies the whole skill folder into ~/.claude/skills/skills-installer-stack-cairn/ — the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.

    Faster alternative (instruction-only skills)

    Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.

    Quick install (SKILL.md only)
    Sign up to copy
  3. Restart Claude Code.

    Quit and reopen Claude Code (or any other agent that loads from ~/.claude/skills/). New skills are picked up on startup.

  4. Just ask Claude.

    Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.

Prefer to read the source first? Open on GitHub.

When Claude uses it

Install, list enabled, validate, or package LiveAgent skills. Use when you need to inspect the skills enabled in the current conversation, import a local skill directory or package, search/install from ClawHub, install from a GitHub repo/tree URL, or reconcile conflicts during an upgrade.

What this skill does

Skills Installer

Install LiveAgent skills into the fixed runtime root while preserving existing user data.

Workflow

  1. Determine the source type: ClawHub slug/search result, local skill directory, .zip or .skill archive, HTTP(S) download, or GitHub URL.
  2. Determine the target destination. Runtime installs always go through SkillsManager into LiveAgent's fixed user Skills root.
  3. For local workspace sources, pass workspace-relative paths such as ./my-skill or ./dist/my-skill.skill; LiveAgent resolves them against the current chat workspace before installing into the fixed skills root.
  4. Read references/install-sources.md and references/safety-and-conflicts.md through file tools before replacing anything. Use path="skill://skills-installer/references/install-sources.md" and path="skill://skills-installer/references/safety-and-conflicts.md", or use the exact pathRef returned by a prior tool.
  5. Use SkillsManager with action=list to inspect the skills enabled in the current conversation when conflict or inventory context matters.
  6. Use SkillsManager with action=clawhub_search to find ClawHub skills when the user asks to search the public store, then use action=clawhub_install with the returned slug to install from ClawHub.
  7. Use SkillsManager with action=install to import a direct local, archive, HTTP(S), or GitHub source. Prefer conflict=backup unless the user explicitly accepts replacement.
  8. After installation, the new skill is enabled for the current conversation automatically; ask LiveAgent to rescan skills or reopen the Skills menu only if the new skill does not appear immediately.

Commands

SkillsManager(action=list)
SkillsManager(action=clawhub_search, query=memory, limit=10)
SkillsManager(action=clawhub_search, sort=downloads, limit=10)
SkillsManager(action=clawhub_install, slug=example-skill, conflict=backup)
SkillsManager(action=install, source=./my-skill)
SkillsManager(action=install, source=./dist/my-skill.skill)
SkillsManager(action=install, source=https://github.com/owner/repo/tree/main/skills/my-skill)
SkillsManager(action=validate, name=my-skill)
SkillsManager(action=package, name=my-skill)

Rules

  • Do not install into the runtime skills root unless the user asked for runtime activation.
  • Do not delete backups automatically.
  • Treat a source as valid when it contains SKILL.md, skill.md, skill.json, or a fallback README.md.
  • When inspecting or maintaining enabled installed skill files, use Read, List, Glob, Grep, Write, Edit, or Delete with the exact path you see: prefer skill://<baseDir>/... or a pathRef returned by a tool. Do not use Bash for workspace or Skill file operations.
  • Use conflict=fail for dry safety checks, conflict=backup for upgrades, and conflict=overwrite only when the user explicitly accepts data replacement.

Related skills