A pack of 24 structured workflow skills and 8 slash commands that guide AI coding agents through define, plan, build, verify, review, and ship phases. It enforces engineering discipline by mapping incoming work to multi-step processes via supported agent CLIs.
Project overview
It provides structured workflows with steps and checkpoints rather than generic reference prose, addressing the tendency of AI coding agents to skip specs, tests, and security reviews.
Project type
AI Coding
Use cases
Coding & Development
Deployment
Refer to project documentation
License
MIT
Best for
Developers and AI engineers using compatible agent CLIs who need to enforce multi-step lifecycle workflows.
Teams looking to counter AI agent tendencies to skip specifications, tests, or security reviews via structured checklists and anti-rationalization prompts.
Key capabilities
A pack of 24 structured workflow skills covering define, plan, build, verify, review, and ship phases, plus a meta-skill to map incoming work to the right workflow.
Provides 8 slash commands that map to the development lifecycle, activating the right skills automatically.
Quick-reference material that skills pull in when needed.
Limitations and risks
Requires an AI agent that accepts system prompts or instruction files.
GPU requirements, hardware minimums, data boundaries, cost dependencies, and telemetry specifics are not documented.
Getting started
Installation is rated easy because the project provides a single command, npx skills add addyosmani/agent-skills, to install everything. No coding is required for setup.
Alternatives and comparisons
Enforces test-driven development and uses a subagent-driven development methodology for coding agents.
Provides over 260 reusable skills, rules, and command shims across multiple language ecosystems and agent harnesses.
Enforces software engineering fundamentals like test-driven development, domain modeling, and code review for coding agents.