
How to use Claude
You'll have read what not to paste into Claude and checked your workplace's AI rules, signed in to Claude (a free account is enough), had a two-turn exchange with it on something safe from your own work, and checked one claim in its answer against the source.
What you'll make
First Claude Session Log
- What I checked before starting: workplace AI rules and what not to paste in
- My prompt: the two-turn exchange I had with Claude on a real work topic
- The claim I verified: what Claude said, what the source said, whether they matched
- What I noticed: where Claude was useful, where I had to push back or correct it
Write this as a plain document — a few sentences per section is enough. You are recording what actually happened, not writing a polished report.
The 4 parts
Each part ends with a check you can see.
Check the Rules Before You Start
Find out what your workplace says about using AI tools, and learn exactly what kinds of content you should never paste into Claude
Done when
You can name at least one specific thing your workplace allows or restricts around AI use, AND you can list at least three categories of content that should never go into Claude — with a concrete example of each
Have Your First Real Conversation
Use Claude's three-part prompt structure to ask about something from your actual work, then push back with a follow-up turn to get a more useful answer
Done when
You have a written record of a two-turn Claude exchange on a real work topic, with your first prompt visibly using the stage-task-rules structure, and your second turn asking Claude to adjust something specific in its first response
Check One Claim Against a Source
Pick the single most important factual claim in Claude's response, find an external source that either confirms or contradicts it, and record what you find
Done when
You have written down: the exact claim Claude made, the source you checked it against (with enough detail that someone else could find it), whether the source confirmed or contradicted the claim, and one sentence about what this means for how you'll use Claude going forward
Write Up What You Learned
Assemble your session log — what you found out before starting, what you asked and got, what you verified, and what you'd do differently next time
Done when
You have a complete four-section session log that documents your actual first Claude session, with specific observations in each section — not generic reflections
Capstone: you put together your First Claude Session Log.
Where this helps
An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.
About this lesson
What you'll be able to do
- Sign in to Claude and have a real two-turn conversation on something from your own work
- Identify at least one claim in Claude's answer and check it against an external source
- Know what kinds of content are safe to paste into Claude and what to leave out
- Write a prompt using a simple three-part structure and get a more useful response than your first try
Who it's for
- Anyone who has never opened Claude before or has only tried it once
- Professionals who want a safe, structured first session with AI — no technical background needed
- People who have heard about Claude at work and want to know how to start without making a mistake
What it won't cover
- Advanced features like Projects, Cowork, Skills, or Connectors — those come later in Claude 101
- Prompt engineering as a craft — this is about your first working conversation, not perfecting prompts
- Team or Enterprise plan setup — this lesson assumes a free or personal account
- Comparing Claude to other AI tools
Skills you'll practise
- Using Claude to have a productive two-turn conversation on a real work task
- Fact-checking Claude's output against an external source
- Recognising what is and isn't safe to paste into Claude
- Structuring a prompt with stage, task, and rules