Agentic Coding

Agent Skills is a production-grade workflow system for AI coding agents that encodes senior engineering practices across the full development lifecycle. It provides 25 structured skills organized into six phases—DEFINE, PLAN, BUILD, VERIFY, REVIEW, SHIP—with nine slash commands that automatically activate relevant skills based on context. Each skill is a step-by-step workflow with verification gates and anti-rationalization tables that prevent agents from skipping specs, tests, security reviews, and quality checks. Key features include /spec for PRD-driven development, /plan for atomic task breakdown, /build for autonomous implementation after single approval, /test for test-driven development, /review for five-axis code quality gates, and /ship for production deployment. The system works across 70+ agents including Claude Code, Cursor, Cline, GitHub Copilot, and others, installable via npx or native integrations. Skills embed best practices from Google's engineering culture—including Hyrum's Law for API design, test pyramid principles, change sizing norms, and trunk-based development—designed specifically for solo founders and small teams using AI agents daily to maintain production-quality standards while working efficiently.

Agentic Product Management Mastery offers an educational framework for product managers to leverage AI agents in workflows, including structured learning modules on requirement definition and PRD drafting specifically tailored for AI agent systems. The resource guides users through autonomous workflow orchestration, multi-agent coordination between human PMs and coding agents, and provides canonical best practices for designing PRDs and workflow architectures compatible with agentic development tools.