Claude Code Complete Course

This course is a full practical path from first install to team rollout. Every module uses official documentation and release sources, with direct links for verification.

Official Sources Used Throughout

• Claude Code overview: https://docs.anthropic.com/en/docs/claude-code/overview
• Claude Code quickstart: https://docs.anthropic.com/en/docs/claude-code/quickstart
• Claude Code tutorials: https://docs.anthropic.com/en/docs/claude-code/tutorials
• Claude Code CLI reference: https://docs.anthropic.com/en/docs/claude-code/cli-reference
• Claude Code settings: https://docs.anthropic.com/en/docs/claude-code/settings
• Claude Code output styles: https://docs.anthropic.com/en/docs/claude-code/output-styles
• Claude Code memory: https://docs.anthropic.com/en/docs/claude-code/memory
• Claude Code MCP: https://docs.anthropic.com/en/docs/claude-code/mcp
• Claude Code SDK MCP: https://docs.anthropic.com/en/docs/claude-code/sdk/sdk-mcp
• Claude Code GitHub Actions: https://docs.anthropic.com/en/docs/claude-code/github-actions
• Claude Code costs: https://docs.anthropic.com/en/docs/claude-code/costs
• Claude Code security: https://docs.anthropic.com/en/docs/claude-code/security
• Anthropic news and release updates: https://www.anthropic.com/news
• Claude Code Action repository: https://github.com/anthropics/claude-code-action
• GitHub Actions docs: https://docs.github.com/en/actions
• GitHub Actions security hardening: https://docs.github.com/en/actions/security-guides/security-hardening-for-github-actions

Course Outcomes

By the end of this course, you will be able to:

1. Install and configure Claude Code for safe daily use.
2. Write deterministic prompts that reduce rework.
3. Run code changes with explicit review gates.
4. Integrate MCP tools with least privilege.
5. Automate PR workflows with GitHub Actions.
6. Track and optimize token costs.
7. Implement team governance for AI-assisted coding.

Module 1 - Setup and First Run

What You Learn

• Installation flow and environment checks.
• Authentication and first interactive session.
• Basic command lifecycle and safe editing posture.

Exercises

1. Install Claude Code and verify command availability.
2. Run your first session in a sandbox repository.
3. Perform one small refactor and inspect the diff.

Screenshot Checklist

• Terminal showing successful install.
• First claude launch.
• Login complete state.
• First proposed diff with approval prompt.

Primary Reading

• Quickstart: https://docs.anthropic.com/en/docs/claude-code/quickstart
• Overview: https://docs.anthropic.com/en/docs/claude-code/overview

Module 2 - Prompt Engineering for Code Tasks

What You Learn

• Constraint-first prompting.
• File scope limits and acceptance criteria.
• Plan then patch then test pattern.

Prompt Template

Objective: [exact outcome]
Constraints: [files allowed, style rules, non-goals]
Process: propose a plan first, then patch, then run tests
Validation: list tests run and summarize risk

Exercises

1. Convert a vague prompt into a constrained prompt.
2. Compare results across three prompt variants.
3. Produce a reusable prompt template library for your team.

Primary Reading

• Tutorials: https://docs.anthropic.com/en/docs/claude-code/tutorials
• Prompt engineering overview: https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering/overview

Module 3 - Diff Quality and Review Discipline

What You Learn

• Breaking large changes into staged commits.
• Review-first behavior before applying broad edits.
• Human review checklist for correctness and maintainability.

Review Checklist

• Does every changed file map to the requested scope?
• Are tests added or updated where behavior changed?
• Is error handling preserved or improved?
• Is rollback straightforward if production issues appear?

Primary Reading

• CLI reference: https://docs.anthropic.com/en/docs/claude-code/cli-reference
• Security docs: https://docs.anthropic.com/en/docs/claude-code/security

Module 4 - Settings, Memory, and Output Control

What You Learn

• Configure output style by task type.
• Use memory features for long-running workflows.
• Reduce context noise during focused implementation.

Exercises

1. Create two settings profiles: concise and teaching.
2. Run the same task with each profile and compare outcomes.
3. Document when each profile should be used.

Primary Reading

• Settings: https://docs.anthropic.com/en/docs/claude-code/settings
• Output styles: https://docs.anthropic.com/en/docs/claude-code/output-styles
• Memory: https://docs.anthropic.com/en/docs/claude-code/memory

Module 5 - MCP Integration Basics

What You Learn

• MCP architecture and trust boundaries.
• Connecting tools safely.
• Diagnosing tool timeout and data-shape failures.

Exercises

1. Configure one MCP server in a test project.
2. Execute one tool-assisted coding task.
3. Validate fallback behavior for tool failures.

Primary Reading

• MCP docs: https://docs.anthropic.com/en/docs/claude-code/mcp
• SDK MCP: https://docs.anthropic.com/en/docs/claude-code/sdk/sdk-mcp
• MCP GitHub org: https://github.com/modelcontextprotocol

Module 6 - MCP Advanced Workflows

What You Learn

• Multi-tool sequencing patterns.
• Stable intermediate outputs.
• Failure handling and retries.

Exercises

1. Implement two-step tool workflow with validation between steps.
2. Add bounded retries and fallback handling.
3. Write an operational runbook for the workflow.

Primary Reading

• MCP TypeScript SDK: https://github.com/modelcontextprotocol/typescript-sdk
• MCP Python SDK: https://github.com/modelcontextprotocol/python-sdk

Module 7 - GitHub Actions Integration

What You Learn

• Action workflow design for pull requests.
• Permissions minimization.
• Secret handling and protected branches.

Exercises

1. Configure anthropics/claude-code-action@v1 in a repo.
2. Trigger review workflow from PR comments.
3. Add timeout, concurrency, and permission limits.

Primary Reading

• Claude Code Actions docs: https://docs.anthropic.com/en/docs/claude-code/github-actions
• Action repository: https://github.com/anthropics/claude-code-action
• GitHub Actions docs: https://docs.github.com/en/actions
• Security hardening: https://docs.github.com/en/actions/security-guides/security-hardening-for-github-actions

Module 8 - Cost Engineering

What You Learn

• Cost drivers in coding sessions.
• Task decomposition for lower token usage.
• Repeatable cost benchmarking.

Exercises

1. Run baseline task and record cost.
2. Apply scope and prompt optimizations.
3. Compare cost and quality before and after.

Primary Reading

• Costs docs: https://docs.anthropic.com/en/docs/claude-code/costs

Module 9 - Security and Governance

What You Learn

• Risk tiers for AI-assisted changes.
• Human review requirements by tier.
• Sensitive data handling boundaries.

Governance Policy Starter

• Tier 1 low risk: docs and non-critical refactors.
• Tier 2 medium risk: feature edits requiring full tests.
• Tier 3 high risk: auth, billing, infra changes with mandatory senior review.

Primary Reading

• Security docs: https://docs.anthropic.com/en/docs/claude-code/security
• Anthropic news for updates: https://www.anthropic.com/news

Module 10 - Team Rollout Plan

What You Learn

• Pilot design and success metrics.
• Change management for engineering teams.
• Standard operating procedures for daily use.

Rollout Framework

1. Week 1: two-engineer pilot.
2. Week 2: evaluate quality and cycle-time.
3. Week 3: expand to one full squad.
4. Week 4: publish org standards and templates.

Module 11 - Production Incident Scenarios

What You Learn

• Detecting incorrect automated edits.
• Rollback and remediation paths.
• Communication templates for incident response.

Exercises

1. Simulate flawed patch in staging.
2. Run rollback with audit notes.
3. Document root cause and prevention controls.

Module 12 - Capstone

Capstone Brief

Build a full feature with this flow:

1. Define acceptance criteria.
2. Generate plan.
3. Apply staged changes.
4. Run tests and lint.
5. Submit PR with risk and rollback summary.
6. Run CI assistant checks and finalize review.

Capstone Scoring

• Correctness: 30 percent
• Code quality: 20 percent
• Test quality: 20 percent
• Security and governance: 15 percent
• Cost discipline: 15 percent

Required Screenshots for Publication

Capture these and add to your course assets folder:

1. Install command and success output.
2. First authentication flow complete state.
3. First plan response.
4. Approval prompt before patch.
5. Diff preview.
6. Test run output.
7. MCP configuration example.
8. MCP tool call result.
9. GitHub Actions YAML excerpt.
10. PR comment trigger example.
11. Action run summary.
12. Cost output comparison.
13. Security checklist file.
14. Capstone final PR summary.

Author QA Checklist

• Every claim includes at least one official link.
• Every lesson includes a hands-on exercise.
• Every module includes at least one screenshot requirement.
• Every advanced module includes cost and risk notes.
• Every workflow can be run in a clean repository from scratch.

Suggested Publishing Plan for Developers Digest

1. Publish this complete guide first.
2. Split each module into individual course lessons in /courses.
3. Add one hero image for the course page at /public/images/courses/.
4. Add a companion blog post for each advanced module.
5. Link all assets from tutorials and guides index pages.

Release Maintenance Cadence

Before each cohort or major promotion:

• Re-check all official docs and release pages.
• Re-run every command shown in lessons.
• Re-capture screenshots if UI or workflow changed.
• Update lesson notes with dated verification.