Reusable Claude Skill Patterns for Prompts and Tools
Code-level patterns for Claude Agent Skills: contract/body split, role-inputs-procedure-stop, the script boundary, and context layering for reusable skills.
Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Code-level patterns for Claude Agent Skills: contract/body split, role-inputs-procedure-stop, the script boundary, and context layering for reusable skills.
Step-by-step guide to building a production Claude Agent Skill — frontmatter, procedural body, scripts, reference files, and testing discovery and execution.
Inside the architecture of Claude Agent Skills — folder layout, the metadata index, discovery, on-demand loading, and how skills compose with MCP tools.
Build autonomous AI shopping assistants for e-commerce with product recommendations, order tracking, returns, and voice shopping.
Build FinTech AI agents for KYC identity verification, document validation, transaction monitoring, and AML compliance workflows.
Build predictive maintenance AI agents for manufacturing with sensor monitoring, anomaly detection, maintenance scheduling, and parts ordering.
Master prompt version control, A/B testing, rollback, and analytics for agentic AI systems running at scale in production environments.
Build multi-agent AI systems for real estate with property search, suburb intelligence, mortgage tools, and viewing schedulers.
Step-by-step guide to building reusable tool libraries for agentic AI with design patterns, validation, testing, and permission systems.