# Oh My Skills

> Use when the user wants to install, launch, inspect, inventory, create, edit, archive, deduplicate, or manage local Codex, Claude, or Agent Skills with the oh-my-skills GUI or CLI.

- **Type:** Skill
- **Install:** `agentstack add skill-aijunja-oh-my-skills-oh-my-skills`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [AIjunja](https://agentstack.voostack.com/s/aijunja)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [AIjunja](https://github.com/AIjunja)
- **Source:** https://github.com/AIjunja/oh-my-skills

## Install

```sh
agentstack add skill-aijunja-oh-my-skills-oh-my-skills
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# oh-my-skills

oh-my-skills manages local `SKILL.md` folders across Codex, Claude, and generic Agent Skills. It provides a local browser GUI plus CLI reports for skill inventory, duplicate detection, safe editing, creation, and archiving.

## Quick Workflow

1. If this repository is the current workspace, run commands from the repository root.
2. If the user only has the skill installed, use the skill folder as the repository root.
3. When the user invokes `$oh-my-skills` with a broad request such as "show my skills", "open the dashboard", "manage skills", or "Codex/Claude skills", launch the GUI dashboard. Do not replace the dashboard with a long text list unless the user explicitly asks for a text report.
4. Use the CLI only for reports, automation, duplicate checks, or when the user specifically asks for terminal output.
5. Keep generated inventory files private unless the user explicitly wants to publish them.

## Launch The GUI

From the repository root:

```powershell
python -m skill_ledger gui
```

If the package is installed:

```powershell
oh-my-skills gui
```

The GUI defaults to `http://127.0.0.1:8765/` and auto-increments the port if it is busy. It supports English/Korean UI and light/dark themes.

## CLI Commands

```powershell
python -m skill_ledger scan --format markdown --out inventory\skills.md
python -m skill_ledger scan --format json --out inventory\skills.json
python -m skill_ledger doctor
```

Use forward slashes instead of backslashes on macOS/Linux:

```bash
python -m skill_ledger scan --format markdown --out inventory/skills.md
python -m skill_ledger scan --format json --out inventory/skills.json
python -m skill_ledger doctor
```

## Install As A Skill

Codex:

```bash
git clone https://github.com/AIjunja/oh-my-skills.git ~/.codex/skills/oh-my-skills
```

Claude:

```bash
git clone https://github.com/AIjunja/oh-my-skills.git ~/.claude/skills/oh-my-skills
```

Windows PowerShell:

```powershell
git clone https://github.com/AIjunja/oh-my-skills.git "$env:USERPROFILE\.codex\skills\oh-my-skills"
git clone https://github.com/AIjunja/oh-my-skills.git "$env:USERPROFILE\.claude\skills\oh-my-skills"
```

## Safety Rules

- Treat `codex-system` and `plugin-cache` skills as read-only.
- Editable user targets are `codex-user`, `claude-user`, and `agents-user`.
- Save operations create a backup under `~/.oh-my-skills/backups`.
- Archive operations move folders to `~/.oh-my-skills/archive`; they do not permanently delete files.
- Keep generated inventory files private unless the user explicitly reviews them for sensitive paths and prompt content.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [AIjunja](https://github.com/AIjunja)
- **Source:** [AIjunja/oh-my-skills](https://github.com/AIjunja/oh-my-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-aijunja-oh-my-skills-oh-my-skills
- Seller: https://agentstack.voostack.com/s/aijunja
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.