What is add-skill?

add-skill is a CLI tool for managing Agent Skills—intelligent artifacts that extend your AI coding agent's capabilities.

npx @dataguruin/add-skill dataguruin/agent-skills

┌   skills 
│
◇  Source: https://github.com/dataguruin/agent-skills.git
│
◆  Repository cloned
│
●  Found 3 skills
│
◆  Select skills to install
│  ◻ coinpika-commit-formatter
│  ◻ coinpika-doctrine-pack
│  ◻ coinpika-pr-reviewer
│
◇  Installed 3 skills ──────────╮
│                               │
│  ✓ coinpika-commit-formatter  │
│  ✓ coinpika-doctrine-pack     │
│  ✓ coinpika-pr-reviewer       │
│                               │
├───────────────────────────────╯
│
└  Done!

Skills are data, not code—they're read by your agent to enhance its behavior without executing on your system.

Installation

# Using npx (recommended)
npx @dataguruin/add-skill <org/repo>

# Global install
npm install -g @dataguruin/add-skill

# Or with pnpm
pnpm add -g @dataguruin/add-skill

Requires Node.js 18+

Quick Start

Install skills from GitHub

npx @dataguruin/add-skill dataguruin/agent-skills

Interactive prompts let you select which skills to install.

List installed skills

npx @dataguruin/add-skill list

┌   add-skill 
│
●  Location: ~/.agent/skills
│
◇  Installed Skills (3) ─────────────────────────╮
│                                                │
│  ✓ coinpika-commit-formatter v1.0.0 (10.1 KB)  │
│  ✓ coinpika-doctrine-pack v1.0.0 (137.9 KB)    │
│  ✓ coinpika-pr-reviewer v1.0.0 (9.1 KB)        │
│                                                │
├────────────────────────────────────────────────╯
│
└  Done!

Check health

npx @dataguruin/add-skill doctor

┌   add-skill 
│
◐  Running health check...
│
◇  Health Check Results ─────────────────────────╮
│                                                │
│  ✓ coinpika-commit-formatter: healthy          │
│  ✓ coinpika-doctrine-pack: healthy             │
│  ✓ coinpika-pr-reviewer: healthy               │
│                                                │
├────────────────────────────────────────────────╯
│
●  Errors: 0, Warnings: 0
│
└  Done!

Commands

| Command | Description | |:--------|:------------| | <org/repo> | Install skills from GitHub repository | | <org/repo#skill> | Install a specific skill | | list | List installed skills | | uninstall <skill> | Remove a skill | | update <skill> | Update to latest version | | doctor | Check integrity and health | | verify | Verify checksums | | validate [skill] | Antigravity compliance check | | analyze <skill> | Deep structure analysis | | lock | Generate skill-lock.json | | cache info | Show cache info | | cache clear | Clear local cache | | init | Initialize skills directory |

Flags

| Flag | Description | |:-----|:------------| | --global, -g | Install to global scope | | --yes, -y | Skip confirmation prompts | | --verbose, -v | Show detailed output | | --json | Output as JSON | | --strict | Exit with error on violations | | --fix | Auto-fix issues where possible | | --dry-run | Preview changes without applying |

Features

Progressive Disclosure

Skills follow a structured hierarchy for efficient agent loading:

skill/
├── SKILL.md              # Core instructions (always loaded)
├── resources/            # Reference materials (on-demand)
├── examples/             # Implementation patterns
├── scripts/              # Helper utilities
├── constitution/         # Governing rules
└── doctrines/            # Behavioral policies

Antigravity Compatibility Score

npx @dataguruin/add-skill analyze coinpika-doctrine-pack

┌   add-skill 
│
●  Path: ~/.agent/skills/coinpika-doctrine-pack
│
◇  Skill Analysis: coinpika-doctrine-pack ───────╮
│                                                │
│  SKILL.md Frontmatter:                         │
│    Name: coinpika-doctrine-pack                │
│    Description: Constitutional governance...   │
│    Tags: coinpika, doctrine, governance        │
│                                                │
│  Structure:                                    │
│    ✓ resources                                 │
│    ✓ examples                                  │
│    ✓ scripts                                   │
│    ✓ constitution                              │
│    ✓ doctrines                                 │
│                                                │
│  Antigravity Score: 100/100                    │
│                                                │
├────────────────────────────────────────────────╯
│
└  Done!

Security by Design

| Feature | Description | |:--------|:------------| | No Execution | Skills are data, never executed on your system | | SHA-256 Merkle Hash | Every skill is cryptographically verified | | Integrity Checks | doctor and verify detect tampering | | Audit Trail | Full provenance tracked in .skill-source.json | | Backups | Automatic backup before destructive operations |

Skill Spec Format

Skills require a SKILL.md file with YAML frontmatter:

---
name: my-skill
description: Brief description for semantic routing
version: 1.0.0
author: your-name
tags: javascript, testing, react
---

# My Skill

Instructions for the agent to follow...

Compatibility

Works with any Antigravity-compatible coding agent:

• Gemini CLI / Antigravity
• Claude Code
• Cursor
• OpenCode
• Cline
• And more...

License

MIT © DataGuruIn