Setup Guide — Browser
No installation needed. Claude runs entirely at claude.ai — you just need a browser and an email address. This guide covers account setup and how to personalise Claude using custom instructions and Projects.
Go to claude.ai and click "Start for free." Sign up with Google or with any email address. If you use email, check your inbox for a verification link and click it to activate your account.
The free plan gives you a limited number of messages per day, access to Claude's standard model, and the ability to upload files to conversations. It is enough to get started and test everything on this site. The paid plan (Claude Pro, around $20/month) gives higher limits and access to the most capable model.
You will see a text box in the centre of the screen. Click it and type a message. Press Enter to send. Try: "What is the WHO definition of health and how has it evolved since 1948?" — a simple test to confirm everything is working.
The browser version of Claude doesn't have access to files on your computer the way Cowork does — but you can still personalise it effectively using two features: Custom Instructions and Projects.
Two features give Claude permanent context about who you are and how you work — without you having to repeat yourself in every session.
Found in Settings → Profile. A text field where you tell Claude about your role, your context, and how you want it to respond. The equivalent of ABOUT ME.md and WRITING RULES.md in one place. Claude applies these instructions to every conversation.
A Projects feature lets you create a dedicated workspace for a specific topic or ongoing task. You can add instructions specific to that project, upload reference documents, and keep all related conversations in one place. Think of each Project as a context-aware workspace.
Go to Settings → Profile → Custom Instructions. You will see a text box. Here is an example of what to write — edit it to match your own situation:
Click Projects in the left sidebar, then New Project. Name it something like "Public Health Analysis" or "WHO Briefings." Inside the project you can:
Every AI platform has usage limits — either message caps (free tier), rate limits, or context window constraints. Rather than treating this as a frustration, understand it as a design feature that encourages better practice.
The principle is simple: working within constraints forces you to structure your work more deliberately. Instead of asking an AI to "do everything," you break the task into discrete pieces, provide clear context, and get focused outputs. This usually produces better results than asking for everything at once.
For specific guidance on your chosen platform's limits and how to optimise within them, refer to the platform's official documentation or resources like Claude 101 which cover advanced usage patterns.
AI will tend to validate your thinking. It won't push back unless you ask it to. You need to deliberately create space for critical thinking.
In your Custom Instructions, include something like: "When I ask for analysis or advice, challenge my assumptions. Tell me if the framing seems wrong, if I'm missing something obvious, or if the evidence points elsewhere. Don't just agree with me."
This is particularly important in health work where decisions affect real people. You want a thinking partner, not a rubber stamp.
There is no point in spending hours perfecting the perfect prompt. The more effective approach is to build good systems and then use simple prompts.
A good system means: clear personalisation files (who you are, how you work), organised context (reference documents, frameworks), and structured workflows (file naming, folder organisation, consistent question-asking patterns). Once these foundations are in place, your prompts can be short and direct because the AI already understands the context.
Build the system first. The prompts will be better and simpler as a result.