Uploaded January 2026 | Updated September 2026, 2 weeks ago
Enroll now in Agent Skills with Anthropic: bit.ly/469a15J
In this course, built in collaboration with Anthropic and taught by Elie Schoppik, Head of Technical Education at Anthropic.
This course shows how to make AI agents more reliable, reusable, and easier to scale by equipping them with agent skills –structured folders of instructions that package expert workflows and load on demand. Instead of repeatedly re-prompting agents or managing bloated context, you define skills once and reuse them across agents and environments using an open standard format.
This course covers how to design, build, and apply agent skills across real workflows for coding, data analysis, research, and many other practical scenarios. You’ll work hands-on with Claude.ai, Claude Code, the Claude API, and the Claude Agent SDK, and see how skills fit alongside tools, MCP, and subagents in modern agentic systems.
You’ll learn how to:
- Turn general-purpose agents into specialists when needed
- Structure skill folders and SKILL.md files for efficient context management
- Use pre-built Anthropic skills and create custom ones following best practices
- Combine skills with MCP and subagents for multi-step workflows
- Build code generation, review, testing, and research agents with isolated context
You’ll leave with concrete patterns for making agents more consistent, reusable, and easier to work with.
Learn more: bit.ly/469a15J
Enroll now in Agent Skills with Anthropic: bit.ly/469a15J
In this course, built in collaboration with Anthropic and taught by Elie Schoppik, Head of Technical Education at Anthropic.
This course shows how to make AI agents more reliable, reusable, and easier to scale by equipping them with agent skills –structured folders of instructions that package expert workflows and load on demand. Instead of repeatedly re-prompting agents or managing bloated context, you define skills once and reuse them across agents and environments using an open standard format.
This course covers how to design, build, and apply agent skills across real workflows for coding, data analysis, research, and many other practical scenarios. You’ll work hands-on with Claude.ai, Claude Code, the Claude API, and the Claude Agent SDK, and see how skills fit alongside tools, MCP, and subagents in modern agentic systems.
You’ll learn how to:
- Turn general-purpose agents into specialists when needed
- Structure skill folders and SKILL.md files for efficient context management
- Use pre-built Anthropic skills and create custom ones following best practices
- Combine skills with MCP and subagents for multi-step workflows
- Build code generation, review, testing, and research agents with isolated context
You’ll leave with concrete patterns for making agents more consistent, reusable, and easier to work with.
Learn more: bit.ly/469a15J
