Skilljar is a learning management system that hosts our educational content. You're logging into it to access the Anthropic course materials. This separate platform allows us to provide interactive learning experiences, track your progress, and ensure you have access to all course resources in an organized way.
Claude Code, MCP & Integration
Learn to make a working Claude integration configurable, shareable, and safe to connect to real systems.
Claude Code is your terminal-native development partner. Building on the prior API foundations, this module covers adding a permission layer, a configuration system, and team-oriented sharing features, it then covers using the MCP protocol to connect Claude securely to external and enterprise systems. Each topic teaches a configuration decision that may feel safe on your own machine but could become a failure when someone else runs it.
Learning objectives
By the end of this module, you will be able to:
- Run Claude Code through the explore, plan, and code loop, and select a permission mode that matches the risk level of the work, so the agent stays productive without being granted more authority than the task requires
- Read AI-generated code, review output with calibrated trust, and place a human review gate where the cost of a wrong call is high
- Give Claude Code durable project context using CLAUDE.md, rules instruction files, hooks, and subagents
- Package a workflow as Skills, custom commands, and a plugin, authoring a Skill once that runs the same way across Claude Code, the Messages API, and the Agent SDK
- Build an MCP server that makes tools, resources, and prompts available to Claude, select the transport that matches how the client and server communicate, and set the configuration scope that controls who loads it
- Connect Claude to enterprise systems, authenticate those connections using patterns a regulated customer will accept, and scope a code modernization engagement so the work holds up under a security review