Most people open Claude, type a question, get an answer, and close the tab.
That’s like buying a Swiss Army knife and only ever using the toothpick.
Anthropic has been quietly building out a proper product suite around Claude, and if you’re only using the chat window, you’re leaving serious productivity on the table. This guide breaks down every major tool, what it does, why it exists, when to reach for it, and where to find it.
🧠 First: The Model Behind It All
Before we get into products, a quick note. The current flagship model is Claude Sonnet 4.6, sitting inside the Claude 4 family alongside Claude Opus 4.6 (the heavier, more capable model) and Claude Haiku 4.5 (the fast, lightweight option).
Think of the model family like engine choices. The tools below are the vehicles those engines sit inside.
💬 Claude Chat (claude.ai)
What it is: The main interface. Available as a web app, mobile app, and desktop app.
What it does: Conversational AI for writing, analysis, research, coding, brainstorming, planning, summarising, and just about anything else you’d want a sharp thinking partner for.
Why it exists: It’s the front door. Most people start here and some never need to go further.
When to use it:
- Drafting emails, documents, posts
- Thinking through problems out loud
- Researching topics and getting structured summaries
- Writing and editing code
- Getting a second opinion on a decision
Standout features you might have missed:
- Memory — Claude can remember things across conversations. You can tell it your role, your preferences, your context, and it carries that forward.
- Projects — Group conversations, upload files, and give Claude persistent context for a specific piece of work. Think of it like a folder with a brain.
- Extended thinking — On complex problems, Claude can be asked to reason through things more carefully before answering.
- Voice — On mobile, you can speak to Claude directly. Useful when you’re moving.
- File uploads — Drop in PDFs, images, spreadsheets, and documents. Claude reads them and works with them.
Where to find it: claude.ai or via the Claude mobile app (iOS / Android).
🖥️ Cowork (Claude in Chrome)
What it is: A browser extension that brings Claude directly into your browsing experience.
What it does: Claude can see what’s on your screen, help you with tasks in real time, and act alongside you as you work in other tools — email, docs, CRMs, whatever you have open.
Why it exists: Most work doesn’t happen in a chat window. It happens in Gmail, Google Docs, Notion, Salesforce, and a hundred other places. Cowork puts Claude where the work actually is.
When to use it:
- You’re reading a long document or email thread and want a quick summary without copy-pasting
- You want Claude to help you draft a reply without switching tabs
- You’re doing research and want Claude to help you synthesise across multiple pages
- You need help with a task inside a tool Claude can see
Where to find it: Chrome Web Store — search “Claude” or look for the Anthropic extension.
⚡ Dispatch
What it is: Claude’s agentic task runner. You give it a goal, it figures out the steps and executes them.
What it does: Dispatch lets Claude handle multi-step tasks on your behalf — researching, compiling, summarising, filing, writing — without you needing to babysit each step.
Why it exists: Single-turn Q&A only goes so far. Some tasks have ten steps. Dispatch lets Claude own the whole thing.
When to use it:
- “Research the top five competitors in this market and give me a comparison table”
- “Go through these emails and pull out every action item”
- “Draft a week’s worth of LinkedIn posts from this briefing doc”
The important bit: Dispatch is still evolving. It’s currently rolling out as part of Anthropic’s broader push into agentic workflows. Keep an eye on the Anthropic product page for the latest.
🖥️ Claude Code
What it is: A command-line tool for developers. Runs in your terminal, not a browser.
What it does: Claude Code integrates directly into your development environment. It can read your codebase, understand its structure, write and edit files, run terminal commands, and execute multi-step coding tasks autonomously.
Why it exists: Copy-pasting code snippets in and out of a chat window is slow and loses context. Claude Code works where developers actually work, with full access to the project.
When to use it:
- Building features across multiple files
- Debugging something that requires understanding the whole codebase
- Refactoring or restructuring code
- Running tests and fixing failures iteratively
Where to find it: Install via npm install -g @anthropic-ai/claude-code — full docs at docs.claude.com.
📊 Claude in Excel and PowerPoint
What it is: Beta integrations that put Claude inside Microsoft Office tools.
What it does:
- Claude in Excel — Helps you build formulas, analyse data, clean up spreadsheets, and explain what’s in a dataset.
- Claude in PowerPoint — Helps you build slides, write content, restructure decks, and improve presentation flow.
Why it exists: A huge amount of knowledge work happens in Office. These integrations reduce the friction of switching between tools.
When to use them:
- You have a messy spreadsheet and need help making sense of it
- You need to build a deck quickly from a set of bullet points or a document
- You want Claude to write the narrative for a presentation rather than just format it
Where to find them: Available as beta products — check claude.ai for rollout updates.
🤖 The API (for Builders)
What it is: Direct programmatic access to Claude models.
What it does: Lets developers integrate Claude into their own products, tools, workflows, and automations. You send Claude a message, it sends one back. At scale, with control.
Why it exists: Not everyone wants a chat UI. Some teams want Claude baked into their own software, customer support systems, internal tools, or automated pipelines.
When to use it:
- You’re building a product that needs AI inside it
- You want to automate a workflow that runs without a human in the loop
- You need fine-grained control over context, system prompts, and model behaviour
Key models via the API:
claude-opus-4-6— Most capableclaude-sonnet-4-6— Smart and efficient, the everyday workhorseclaude-haiku-4-5-20251001— Fast and lightweight for high-volume tasks
Where to start: docs.claude.com
🏢 Claude for Teams and Enterprise
What it is: Upgraded plans for organisations that need more control, more capacity, and better collaboration features.
What it does:
- Higher usage limits
- Centralised billing and admin
- Enhanced privacy and data controls
- Team-level Projects and shared context
Why it exists: Individual limits hit fast when a whole team is using Claude daily.
When to use it: When you’re past the point of managing individual accounts and need Claude to work properly at an organisational level.
Where to find it: claude.ai/upgrade
🗺️ Quick Reference: Which Tool When?
| What you’re doing | Tool to reach for |
|---|---|
| Writing, research, thinking | Claude Chat |
| Working inside your browser | Cowork (Chrome extension) |
| Autonomous multi-step tasks | Dispatch |
| Coding across a full codebase | Claude Code |
| Building spreadsheets or decks | Claude in Excel / PowerPoint |
| Building your own AI product | The API |
| Running Claude across a whole team | Teams / Enterprise |
The Honest Summary
Most people are using about 10% of what Claude can actually do.
The chat window is useful. The rest of the toolkit is what turns Claude from a novelty into a genuine operating advantage.
Start with Chat. Add Cowork if you live in your browser. Graduate to Claude Code if you build things. When tasks get complex enough that you need Claude to run them rather than just advise on them, that’s what Dispatch is for.
Pick the right tool for the right job. That’s the whole game.
Want sharper prompts to get more out of Claude and every other AI tool? Browse the HackTheSim Prompt Hub — field-tested patterns you can copy and adapt straight away.