The Wow Moment

No. Claude Code installs via a single terminal command and the setup wizard walks you through connecting your Anthropic account. If you can install an app on your phone, you can install Claude Code. We cover every step in this module, including what to do when something goes wrong.

Claude Code runs on macOS, Windows, and Linux. On Windows you will use Git Bash or WSL as your terminal. This module covers the specific setup for each platform so you are not left guessing.

Claude Code uses your Anthropic API credits. Typical business use costs between $20 and $100 per month depending on usage volume. We show you how to set spending limits and monitor your usage so there are no surprises on your bill.

Absolutely. Claude Code runs in your terminal and works alongside any editor or IDE you already use. Many people keep VS Code open for browsing files while Claude Code handles the heavy lifting in a separate terminal window.

CLAUDE.md is a markdown file that lives in your project root and acts as a permanent instruction set for Claude Code. Every time Claude Code starts a conversation in that project, it reads this file first. Think of it as the job description you hand to a new employee on their first day.

Start with 20 to 50 lines covering your business name, brand rules, and two or three non-negotiable constraints. A short, focused CLAUDE.md outperforms a 500-line document full of vague guidance. You will expand it as you learn what Claude Code needs to know.

Nothing breaks permanently. CLAUDE.md is just a text file you can edit at any time. If Claude Code starts producing off-brand work, you update the file and the next conversation picks up the new instructions immediately. Iteration is the whole point.

Yes, and you should. Each project folder gets its own CLAUDE.md tailored to that project's rules, technologies, and constraints. You can also have a global CLAUDE.md in your home directory for rules that apply everywhere, like your brand colours or UK spelling preferences.

Never. CLAUDE.md is typically committed to version control and shared with your team. Store credentials in environment variables or dedicated secrets files that are gitignored. Your CLAUDE.md should reference where secrets are stored, not contain the secrets themselves.

A skill is a reusable set of instructions stored as a markdown file that Claude Code can invoke on demand. Instead of explaining the same task every conversation, you write it once as a skill. For example, a skill might define exactly how to generate a client report or how to write a blog post in your brand voice.

We recommend starting with the five skills that save you the most repetitive time: a content creation skill for your brand voice, a data formatting skill for your reports, a code review or QA skill, a client communication template skill, and a research summarisation skill. This module guides you through choosing based on your specific business.

Test each skill with three different inputs and compare the outputs against what you would have produced manually. If the output consistently matches your quality standard, the skill is ready. If it drifts, the skill instructions need tightening. We cover a structured testing approach in this module.

Yes. Once you have foundational skills in place, you can create higher-level skills that reference them. For example, a monthly report skill might invoke your data analysis skill, then your chart generation skill, then your brand formatting skill in sequence. We cover this composition pattern in Phase 3.

A quick win is any task that takes you one to four hours manually but can be completed by Claude Code in under 30 minutes once set up. Common examples include generating weekly reports, reformatting data exports, drafting standard client communications, or building simple internal tools.

Most businesses complete their first quick-win project within the first week of the programme. If your quick win saves two hours per week, you have recouped the setup time within a fortnight. The module focuses on choosing projects with immediate, measurable time savings.

That is expected and normal at this stage. The quick-win module teaches you a review-and-refine workflow where you check every output before it leaves your desk. You are building the habit of verification that becomes critical in later phases. No output goes to a client without your approval.

That depends on your business context, but we recommend transparency where appropriate. Many businesses find that clients care about the quality of the output, not how it was produced. This module includes guidance on positioning AI assistance professionally without undermining trust.

Prompts are one-off instructions that disappear after a conversation. The Skill Creation Framework gives you a repeatable structure: a SKILL.md file with defined inputs, outputs, constraints, and verification steps. It is the difference between scribbling a note and writing a proper process document.

No. Skills are written in plain English as markdown files. You describe what you want done, what rules to follow, and what the output should look like. Claude Code reads these instructions and executes them. If you can write a clear email, you can write a skill.

Most businesses start with 5 to 10 skills in Phase 1 and grow to 30 to 50 by Phase 3. Some power users end up with over 80. The framework ensures that even large skill libraries remain organised and maintainable rather than becoming a tangled mess.

Writing skills that are too vague. A skill that says 'write a good blog post' will produce inconsistent results. A skill that specifies word count, tone, structure, target audience, and brand voice produces reliable output every time. This module teaches you to be precise without being rigid.

Yes. Skills are just files in your project directory, so they can be committed to version control and shared across your entire team. Everyone gets the same quality of output because everyone is using the same instructions. This is one of the biggest advantages of the framework approach.