A curated collection of Agent Skills by Montimage that supercharge AI agents with specialized workflows, domain expertise, and reusable tooling for software development tasks. Works with Claude Code, Codex, OpenClaw, and other AI tools that support skill-based workflows.

What Are Agent Skills?

Agent Skills are modular, self-contained packages that transform AI agents from general-purpose assistants into specialized tools. Each skill provides procedural knowledge, scripts, reference material, and templates for specific development workflows.

Available Skills

Code Quality & Testing

Skill	Description
code-review	Review code for smells, security issues, and Pragmatic Programmer violations with severity-based reports
test-coverage	Identify untested branches, error paths, and edge cases, then generate targeted tests

DevOps & Releases

Skill	Description
devops-pipeline	Auto-detect project stack and configure pre-commit hooks + GitHub Actions CI
release-notes	Generate categorized release notes from git history, merged PRs, and closed issues

Documentation & Open Source

Skill	Description
docs-generator	Analyze project type and restructure documentation into a coherent hierarchy with diagrams
oss-ready	Add README, CONTRIBUTING, LICENSE, CODE_OF_CONDUCT, SECURITY, and GitHub templates

Infrastructure & Optimization

Skill	Description
install-script-generator	Generate cross-platform install scripts with environment detection, verification, and rollback
ollama-optimizer	Detect hardware (GPU/RAM/CPU), classify tier, and optimize Ollama for maximum performance

Skill Development

Skill	Description
skill-creator	Guided 4-phase skill creation: Discovery, Approval, Build, Test & Deliver
skill-auditor	Scan skills for security risks, prompt injection, and malicious patterns before installation

Quick Start

Installation

Install all skills with a single command:

npx skills add https://github.com/Montimage/skills

Or install a specific skill:

npx skills add https://github.com/Montimage/skills --skill code-review
npx skills add https://github.com/Montimage/skills --skill oss-ready

Usage

Once installed, skills are automatically detected by your AI agent. Simply describe what you need in natural language:

# Triggers the code-review skill
> Review my code for smells and quality issues

# Triggers the oss-ready skill
> Make this project open source ready

# Triggers the release-notes skill
> Generate release notes for the latest version

Project Structure

skills/
├── skills/                        # All agent skills
│   ├── code-review/               # Code quality reviews
│   ├── devops-pipeline/           # CI/CD and pre-commit setup
│   ├── docs-generator/            # Documentation restructuring
│   ├── install-script-generator/  # Cross-platform installers
│   ├── ollama-optimizer/          # Local LLM optimization
│   ├── oss-ready/                 # Open-source readiness
│   ├── release-notes/             # Changelog generation
│   ├── skill-auditor/             # Security audit for agent skills
│   ├── skill-creator/             # Skill authoring guide
│   └── test-coverage/             # Test coverage expansion
├── assets/                        # Logo and screenshot assets
├── docs/                          # Project documentation
│   ├── ARCHITECTURE.md            # System design and decisions
│   ├── CHANGELOG.md               # Version history
│   ├── DEPLOYMENT.md              # Publishing and distribution
│   └── DEVELOPMENT.md             # Local development setup
├── LICENSE                        # Apache 2.0
├── CONTRIBUTING.md                # Contribution guidelines
├── CODE_OF_CONDUCT.md             # Community standards
└── SECURITY.md                    # Vulnerability reporting

Skill Anatomy

Each skill follows a standard structure:

skill-name/
├── SKILL.md          # Required: metadata + instructions for the AI agent
├── README.md         # Required: human-readable documentation
├── scripts/          # Optional: executable code (Python/Bash)
├── references/       # Optional: domain knowledge, documentation
└── assets/           # Optional: templates, icons, fonts

The SKILL.md file contains YAML frontmatter (name, version, description) that the AI agent reads to determine when to activate the skill, plus markdown instructions loaded on activation. The README.md provides human-readable documentation with highlights, trigger phrases, workflow diagrams, and usage instructions.

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines on:

• Adding new skills
• Improving existing skills
• Reporting bugs and requesting features

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.