Grok Bot: a company in 72 hours

Glossary

The words that come up in the course and are worth having clear. Each one links to the lesson that explains it in depth. Inside lessons, terms with a dotted underline open their definition.

Auto-review

The automatic review that decides which of a Bot's actions need your approval before they happen.

Explained in lesson 04
Bot

An AI teammate with a name, a job and its own conversation. It works on a computer in the cloud, even when you're away.

Explained in lesson 01
Chief of staff

A coordinator Bot: it hands work to the other Bots and only reports blockers and decisions to you.

Explained in lesson 13
Cloud agent

A Cursor coding agent that works on its own computer on a copy of the code and returns the change for review.

Explained in lesson 15
Connector

A ready-made link between a Bot and a service such as Slack, Notion or GitHub. In the app they appear as Plugins.

Explained in lesson 08
CRM

The tool where a team tracks customers and sales deals.

Explained in lesson 19
Evals

Test cases run again after every change to check that a Bot still answers correctly.

Explained in lesson 25
Harness

The tools around a model that let it read, run and act. The same model performs differently depending on its harness.

Explained in lesson 15
MCP

A standard that lets an agent use a tool directly through its API, without clicking through a screen.

Explained in lesson 08
Playbook

The Bot team's manual: the rules they all follow, in a shared document.

Explained in lesson 13
Pull request

A proposed change to the code, packaged so someone can review it before it's accepted.

Explained in lesson 15
Routine

Work a Bot does on its own, on a schedule or when something happens in another tool.

Explained in lesson 10
Skill

Saved instructions to do a task the same way every time. All your Bots can use them.

Explained in lesson 06
Teach a task

Recording how you do a task on the Bot's computer so it learns it and saves it as a skill.

Explained in lesson 06
Tokens

The unit that measures the model's work: what it reads and what it writes. It's what uses up your plan.

Explained in lesson 21

Back to the syllabus