beginnerSelf-paced

Course

AI Fluency Starter

Learn the core language of LLMs, AI agents, tools, APIs, MCP, memory, and workflows.

Duration

6 daily challenges

Format

Self-paced

Instructor

Agent Commons

Checking enrolment…

Course promise

What this course helps you do

A free introductory CommonLab skill path based on the Agent Commons AI education slides.

Work through the course in a focused sequence, then turn the lessons into a concrete result you can keep using.

Who this is for

Beginner learners who want practical progress

People who prefer learning by building rather than watching passively

Teams or individuals who want clearer workflows, reusable knowledge, and better execution

What you will work through

01Make sense of AI

02Creating an AI Agent

03Make Your Agent Powerful

04Intro to Multi-Agent Systems

Expected output

<p>Choose one practical agent capability and map the pieces behind it: LLM, memory, workflow, tools, APIs, or MCP.</p>

About this course

AI Fluency Starter turns the Agent Commons introduction into focused daily challenges.

Learners study one concept at a time, view the original visual material first, then take a short quiz that checks understanding without going beyond the source material.

Daily skill badges

Complete focused challenges one at a time and turn practical work into visible, earned skills.

Open skill path

AI map

50 pts

LLM vs agent

60 pts

Core loop

65 pts

Memory + workflows

55 pts

APIs + MCP

60 pts

Agent capabilities

65 pts

Curriculum

Course outline

4 modules22 lessons~182 min total

Let us make sense of AI

AI fluency starts with a simple map. An LLM is the language model layer: it predicts and works through words. An agent is an LLM extended with memory, tools, and workflows so it can work toward a goal. APIs, tool calls, and MCP are connection ideas: they help AI reach software, tools, and external services. Automation and workflows are about repeatable action, so useful tasks can happen in a structured way instead of being done from scratch every time.

Preview4 min

Understanding LLMs and agents

An LLM is like super-autocomplete: it predicts words and can reason through language. That makes it useful for generating, continuing, transforming, and explaining text. An agent is an LLM extended with memory, tools, and workflows so it can act toward goals. The agent is not automatically smarter. It is connected to more structure and capabilities.

Preview5 min

How it works: the core loop

The core loop is how an agent moves from input to result: Think, Act, Observe. Think: the LLM produces the next step in text. Act: the system executes steps using tools. Observe: the result is reviewed. If the result is not ready, the loop can continue. If it is ready, the system returns the result.

Preview5 min

Awareness and consistency

Memory and workflows make agent behaviour less one-off and more consistent. Memory remembers past actions, results, preferences, and current state. Workflows add structure for repeatability and consistency. In simple terms: memory helps the system know what has happened and what matters. Workflows help it follow a reliable shape each time.

Preview4 min

Connecting AI to the real world

APIs are interfaces that connect software to external services. MCP is a standard way to expose APIs and tools to AI agents. So APIs are a general software connection idea, while MCP is specifically about making tools and context available to AI agents in a standard way.

Preview5 min

What AI agents can do for you

AI agents can find and manage information across systems. They can handle repetitive tasks, help plan and make decisions, and take action across different apps. These capabilities come from the earlier pieces working together: an LLM for reasoning through language, memory and workflows for consistency, and APIs, MCP, and tools for connecting to real systems.

Preview5 min