# Skillkit

> Local-first analytics and structural audits for AI agent skills. Use when reviewing skill health, context cost, private local or Tailscale-remote usage receipts, conflicts, coverage, token burn, or pruning.

- Skill: `crafter-station/skillkit` (Agent Skill)
- Install (CLI): `npx skillmds@latest add crafter-station/skillkit`
- Raw SKILL.md: https://api.skillmd.com/api/skills/crafter-station/skillkit/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Data & Analytics
- Author: crafter-station (https://skillmd.com/u/crafter-station)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/crafter-station/skillkit

---


# skillkit

This file is a discovery stub. Load the version-matched workflow from the installed CLI before running skillkit commands:

```bash
skillkit skills get core
skillkit skills get core --full
```

The first command returns the operating workflow. Use `--full` when exact flags, JSON fields, CI behavior, or the complete command catalog matter.

The CLI ships its own skill content, so the instructions match the installed skillkit version. Run `skillkit skills list` to see the bundled catalog.

If `skillkit` is unavailable, install it with Bun:

```bash
bun add -g @crafter/skillkit
```