VORCL
← All courses
Intermediate · 8 lessons

Claude Code: Agentic Development in the Terminal

A complete Claude Code course: from the agentic CLI mental model to headless CI automation — CLAUDE.md, slash commands, hooks, subagents, worktree isolation, and MCP with real examples.

Start →
  1. 01The Agentic CLI Mental ModelHow a terminal agent differs from IDE autocomplete: the action loop, tools, the context window, and the Claude Code permission model.40 min
  2. 02CLAUDE.md and Project MemoryThe CLAUDE.md hierarchy, quick additions via #, @file imports, and context hygiene: how to make the agent remember your team rules.45 min
  3. 03Slash Commands, Hooks, and SkillsExtending Claude Code: custom slash commands in .claude/commands, deterministic hooks in settings.json, and reusable skills.50 min
  4. 04Subagents and Parallel SessionsDelegating to subagents with isolated context and running several sessions in parallel without collisions using git worktree.55 min
  5. 05Worktree-конвейер: фоновые агенты, которые не мешаютПротокол автономной фоновой работы: worktree как дефолт, единый корень и именование, разделение ролей исполнителя и мёржера, уборка как часть definition of done.50 min
  6. 06MCP Servers in Claude CodeModel Context Protocol in practice: connecting servers via claude mcp add, scopes, .mcp.json, and the security of external tools.45 min
  7. 07Headless Mode and CI Automationclaude -p as a unix utility: output formats, permission flags, GitHub Actions, and safe code-automation patterns.50 min
  8. 08Capstone: A Feature from Spec to PRAn end-to-end project: planning, worktree isolation, subagents, evidence-based verification, and a pull request driven by an orchestrated session.60 min

Sources

Claude Code: Agentic Development in the Terminal · VORCL