Claude Code for Beginners: A Hands-On Course
🔒 This course requires registration
To access this course and all our learning materials, please register for the AI Fluency programme.
Register Now →Frequently asked questions
What is Claude Code and how does it differ from a standard chatbot?
Claude Code is a terminal-based AI agent that executes code, edits files, and runs commands directly on your machine. Unlike a standard chatbot that only provides text suggestions, it operates within your local environment to perform actual development tasks. This allows it to modify source code, run tests, and manage git repositories without requiring manual copy-pasting between windows.
How does the permission system in Claude Code work to prevent unwanted changes?
The permission system requires explicit user approval before Claude Code executes potentially destructive commands or modifies files. It categorises actions into safe read-only operations and risky write operations. Users can configure specific rules to allow or deny certain commands automatically. This ensures that the AI cannot silently alter your codebase or run arbitrary scripts without your direct consent and oversight.
What is the purpose of the CLAUDE.md file in a project directory?
The CLAUDE.md file serves as a persistent context document that Claude Code reads at the start of every session. It contains project-specific instructions, coding standards, and architectural notes that guide the AI’s behaviour. By defining these rules, developers ensure the assistant follows established patterns and understands the project’s unique requirements. This reduces the need to repeatedly explain the same context during long development sessions.
Can Claude Code use external tools through the Model Context Protocol?
Yes, Claude Code supports the Model Context Protocol to connect with external tools and data sources. This allows the assistant to interact with databases, APIs, or other services beyond the local file system. Developers can configure MCP servers to provide specific capabilities, such as querying a Jira instance or accessing a cloud storage bucket. This integration extends the agent’s utility beyond simple code editing to broader workflow automation.
What are subagents in Claude Code and when should they be used?
Subagents are specialised AI instances that handle specific tasks in parallel or isolation. They are useful for complex workflows where different parts of a project require distinct contexts or tools. For example, one subagent might analyse a large codebase while another writes unit tests. This parallel processing speeds up development and prevents context window saturation, allowing the main agent to focus on high-level coordination and integration.
