⬡ Internal · Marketing Team · April 2026

Introduction to AI Tools:
What They Are and
When to Use Each

A practical guide to Claude Chat, Artifacts, Cowork,
Skills, Connectors, Agents, n8n, and the API

// mental model

The LLM is the brain. These are the ways to use it.

The further right → the more powerful, the more AI can do for you

— inside Claude —
— outside Claude —
💬
Chat
claude.ai
📄
Artifacts
in chat
🖥️
Cowork
desktop app
🧩
Skills
playbooks
🔌
Connectors
MCP
⚙️
Agents
n8n / Genie
👩‍💻
API
engineering
← you do more AI does more for you →
💬
01 · simplest entry point

Claude Chat

What it is

Browser-based conversation at claude.ai. You type, Claude responds. Supports connectors if set up — so Claude can also pull from Slack, Drive, or Calendar directly.

Best for

  • Drafting copy or emails
  • Summarizing pasted content
  • Brainstorming campaign angles
  • Editing a brief or doc
  • Quick questions

Marketing example

Paste in a product update → ask for three email subject line variations for different audience segments. Or, with a Slack connector, ask Claude to summarize #product-updates this week — without copying anything yourself.

📄
02 · output layer

Artifacts

Artifact
Chat response
Cowork output
Where it lives
Side panel in chat
Conversation thread
File on your machine
Auto-saves?
❌ No — you copy it out
❌ No
✅ Yes
Can iterate on it?
✅ Yes — in real time
Partially
✅ Yes
Works with Cowork?
✅ Yes — Cowork = input, artifact = output

⚠️ Key rule

Artifacts don't auto-save. When done refining, copy it out to Confluence, a doc, or wherever it belongs. Don't assume Claude is saving your work.

🖥️
03 · desktop app

Claude Cowork

What it is

A desktop app that runs on your computer with access to your local files and folders. Point it at a folder, describe a task — Claude works through it without you opening each file.

Best for

  • Summarizing multiple docs at once
  • Turning a strategy doc into a deck
  • Pulling themes across research files
  • Drafting briefs from folder contents

Key difference from chat

Chat works from what you paste in. Cowork accesses your actual machine. No copying, no pasting — just point and describe.

Works with artifacts

Cowork handles the input (your files). Artifacts handle the output (the rendered deliverable you refine). They work together in one session.

🧩
04 · the playbook

Skills

What it is

Pre-packaged instructions that tell Claude how to do a specific task — the process, format, and rules to follow. Anthropic ships some out of the box (PowerPoint, Excel, Word, PDF). Teams can build custom ones.

Why it matters

Without a skill, Claude improvises — good, but inconsistent. With a skill, it follows the same tested process every time. Less cleanup, more predictability.

Skill vs. Artifact — the key distinction

Skill = the playbook Claude follows  ·  Artifact = the thing Claude produces using that playbook. One is input; one is output.

// example: a "campaign brief" skill enforces your team's format every time — no one remembers the template
🔌
05 · connectivity layer

MCP / Connectors

🤖 Claude
— connects via MCP protocol —
💬Slack
📅Google Cal
Asana
📊CRM
📁Drive
🗄️Database

// same concept, different name depending on the tool

Claude → Connectors
ChatGPT → Apps
Others → Plugins / Integrations
All built on MCP standard

🖥️ Local Extensions (Desktop)

You can install connectors yourself — no ops needed. In the Claude desktop app go to Settings → Extensions and drag in a .MCPB or .DXT file, or browse the gallery. Each extension needs an API key from the service it connects to. Most are read-only — always check the docs before assuming it can write or act.

// connectors — security

API Keys: Treat Them Like Passwords

⚠️ When setting up connectors locally, you'll need an API key. Most connectors are read-only — assume this unless confirmed otherwise.

✅ DO

  • Store keys in 1Password
  • Share keys via 1Password share feature only
  • Revoke immediately if exposed — generate a new one
  • Check connector docs for what actions are actually supported
  • Ask ops if you're unsure what a connector can do

❌ NEVER

  • Paste a key into Slack
  • Send a key via email
  • Paste a key into Claude chat or any AI conversation
  • Store in a text file, sticky note, or shared doc
  • Assume a connector can write/act — verify first
⚙️
06 · automation engine

Agents / Workflows

🖥️ Cowork

Multi-step tasks you kick off manually, within a session. Great for one-off complex jobs — research, write, save, send. You trigger it. It runs once.

⚙️ n8n

Anything that needs a schedule, an event trigger, or has to run without you initiating it. More robust, persistent, built for production.

Trigger
schedule / event
📥
Fetch data
Slack, Sheets...
🤖
AI Agent node
Anthropic model
Verify
check output
📤
Output
Slack / doc / email

Claude ≠ Agent

Claude is a product you interact with. An agent is a concept — an AI model running steps autonomously inside a workflow. In n8n, you use an AI agent node powered by an Anthropic model, not Claude as a product. You can't use Claude directly in n8n.

n8n at Databox

Cloud-based, company access provided. AI agent nodes call Anthropic models directly — no chat interface, no Claude product. The model is just one step in a larger workflow.

🧠 Genie

Genie is a Databox feature built on our own agentic platform — separate from n8n. You can trigger it from n8n workflows via the Databox MCP, but Genie itself is not an n8n workflow.

⚠️ Before going live

Always test without activating first. If a workflow touches a live system (Slack, Asana, email, database) → get ops sign-off before activating.

👩‍💻
07 · engineering layer

The API

What it is

The API is the developer interface — how engineers connect to Anthropic models programmatically to build custom tools, products, and integrations. It's the lowest level of access, and the most flexible.

What you can build with it

  • AI embedded in your CMS or internal tools
  • Automated content review pipelines
  • Custom briefing assistants inside your project management tool
  • Anything that doesn't fit an off-the-shelf connector

Your role

You won't use the API directly — that's engineering territory. But knowing it exists means you can propose ideas. "Could we build something that automatically does X?" is a valuable question. Bring it to ops — the API is probably how they'd answer it.

// quick reference

Which tool should I use?

SituationReach for
One-off task, everything fits in a message 💬 Claude Chat
You want a deliverable to refine and copy out 📄 Artifact
Task involves your local files or folders 🖥️ Cowork
Recurring task needing consistent structure 🧩 Skill
Need Claude to read your Slack, Calendar, Asana 🔌 Connector
High-volume, multi-step, or scheduled task ⚙️ Agent / n8n
Build something custom into a product or system 👩‍💻 API → talk to engineering

// the more Claude does for you → the further right on the spectrum

🎁
bonus · power-ups

Plugins

What is a plugin?

A plugin is a pre-packaged bundle that combines connectors, skills, and tools in one installable file. Instead of setting things up one by one, you install a plugin and get everything configured at once.

Think of it like installing an app — but for Claude's capabilities.

How to install

  • In Cowork: Settings → Extensions → drag in a .MCPB or .DXT file
  • Or browse the extension gallery directly from the settings screen
  • Plugins can be Anthropic-built, team-built, or sourced from a plugin marketplace

What's inside a plugin?

  • One or more connectors (MCP tools) for services like Slack, Asana, Google
  • Skills — pre-written instructions for specific tasks
  • Configuration already set up — less manual work for you

Why it matters

Plugins lower the barrier. You don't need to know how MCP works or how to write a skill — you just install the plugin and start using it. As we build more internal plugins, this is how the team will get access to new AI capabilities.

// your homework

Start this week

1
Try Claude chat today — go to claude.ai and run one real work task through it. Drafting, summarizing, brainstorming — anything.
2
Download Cowork — point it at a folder with real files and ask it to do something useful. This alone will change how you work.
3
Identify one thing to automate — pick a recurring task. Write down: what triggers it, what it needs, what it produces. Bring it to ops. You don't need to know how to build it.
questions → operations  ·  full guide + cheatsheet on confluence  ·  april 2026
⬡ Meta · April 2026

This video was made with AI.

The content, the script, the voiceover — all produced with AI.
The only manual step was assembling it in a video editor.

🖥️
CONTENT · SCRIPT · SLIDES
Claude Cowork
🎙️
VOICEOVER
Claude Code  ·  fish.audio API  ·  text-to-speech
🎬
VIDEO EDITING
Wondershare Filmora  ·  manual

// practice what we preach