Working with Claude Code

CLAUDE.md, effective prompting, subagents, MCP, hooks, Git/PR workflow, and security — making Claude Code a daily production tool.

What Claude Code is, and why it differs from an ordinary chat
CLAUDE.md and project configuration
Writing an effective prompt for a coding agent
Plan Mode
Permission modes
Subagents and parallel tasks
MCP — connecting external tools
Hooks for automation
Orchestrating work — Workflows
Commits and branches through the agent
Opening and describing a PR
Code review with AI
Permissions and sandboxing
When to stop and check — self-review before push
Common mistakes and how to avoid them
End to end: plan → implement → PR on a real feature