addyosmani / agent-skills

2026-02-15 · Source: Github Trending: All languages · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Intermediate, long

Summary

Addyosmani's "Agent Skills" is a collection of 24 production-grade engineering skills designed to guide AI coding agents through the software development lifecycle. This open-source project packages senior engineer workflows, quality gates, and best practices, ensuring AI agents consistently apply them from idea refinement to shipping. It features 7 slash commands that correspond to development phases like /spec, /plan, /build, /test, /review, and /ship, with an optional /build auto for autonomous task execution. Each skill provides a structured workflow, verification requirements, and anti-rationalization tables to prevent agents from skipping critical steps. The package also includes 4 specialist agent personas and 4 reference checklists. It supports integration with various platforms such as Claude Code, Cursor, Antigravity CLI, Gemini CLI, GitHub Copilot, and other agents that accept Markdown-based instructions. The skills embed principles from Google's engineering culture, promoting disciplined, production-quality software development.

Key takeaway

For AI Engineers integrating coding agents into their development pipelines, adopting Agent Skills is crucial for enforcing production-grade engineering discipline. This framework ensures your agents follow established best practices, from rigorous specification and incremental building to comprehensive testing and security reviews. You should implement these structured workflows to standardize agent behavior, reduce common errors, and significantly elevate the quality and reliability of agent-generated code, moving beyond prototype-level outputs.

Key insights

AI coding agents can achieve production-grade software quality by following structured, verifiable engineering workflows and best practices.

Principles

• Spec before code
• Tests are proof
• Verification is non-negotiable

Method

AI agents execute structured workflows, activated by 7 slash commands, across define, plan, build, verify, review, and ship phases. Each skill includes steps, checkpoints, anti-rationalization, and evidence requirements.

In practice

• Install via `/plugin marketplace add addyosmani/agent-skills` for Claude Code
• Use `/build auto` for autonomous task execution
• Apply specialist personas like `code-reviewer` for targeted quality gates

Topics

• AI Agents
• Software Engineering
• Development Workflow
• Code Quality
• Test-Driven Development
• CI/CD

Code references

• addyosmani/agent-skills
• addyosmani/agent-skills

Best for: AI Architect, Machine Learning Engineer, AI Engineer, Software Engineer, MLOps Engineer

Related on AIssential

• You Have Never Seen An AI Agent Do This Before (Claude Code) — Autonomous AI agents can perform complex, multi-stage projects from ideation to social media promotion.
• openai / skills — Agent Skills standardize AI agent capabilities for repeatable task execution across various applications.
• Skill issue: Lessons from skilling up coding agents to use Langfuse - Marc Klingen, Clickhouse — Agents with formalized "skills" can solve complex, multi-domain problems more reliably than traditional workflows.
• Beyond Spec-Driven Development: The Agentic Engineering Playbook That’s Replacing How We Build… — Agentic engineering integrates AI agents into software development with structured human oversight for intent definition and output verification.
• I Stopped Coding for 30 Days — AI Agents Built Everything Instead — AI agents can autonomously handle a broad spectrum of software engineering tasks, fundamentally altering development paradigms.
• An AI state of the union: We’ve passed the inflection point, dark factories are coming, and automation timelines | Simon Willison — AI has fundamentally reshaped software engineering, making code generation cheap and amplifying experienced engineers' capabilities.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Github Trending: All languages.