Anthropic's Free Claude Learning Path: All 9 Courses, Beginner to Advanced

Anthropic published a free, zero-paywall learning path for Claude that runs from first conversation to building your own AI systems. The video walks through all nine courses hosted on Anthropic's Skilljar academy. Most people stop at the intro course. The point of the video is that the advanced half of the track -- MCP, agent skills, and subagents -- is where you stop chatting with Claude and start building with it. The frames show the actual course pages, so this writeup maps each course to what it covers and who it is for.

The Path Has Nine Courses, All Free

A frame near the end shows the full list with links, all pointing to anthropic.skilljar.com (two of them are YouTube playlists). The courses in order:

Claude 101
AI Fluency: Framework & Foundations
Claude Cowork for Business Owners
Claude Code in Action
Claude Code for Business Owners
Building with the Claude API
Intro to Model Context Protocol
Intro to Agent Skills
Intro to Subagents

Every course page in the frames is marked FREE, with a "Register | FREE" or "Enroll in Course" button. The branding on the frames reads "FREE · ZERO PAYWALL" and "9 courses."

Courses 1-3: Foundations and Business Use

Claude 101 teaches how to use Claude for everyday work tasks, core features, and where to go for more advanced topics. The curriculum frame shows modules like "Meet Claude" (What is Claude, Your first conversation, Getting better results, Claude desktop app) and "Organizing your work" (Introduction to projects, Creating with artifacts, Working with files).

AI Fluency: Framework & Foundations is aimed at building general AI judgment, not just Claude mechanics. The course page lists 14 lectures, 1.1 hours of video, 1 quiz, and a certificate of completion. The "About this course" copy frames it around responsible AI development and bringing in a range of human perspectives. It is a co-production -- the page references a partnership (the frame shows Anthropic alongside Ringling College and a named instructor).

Claude Cowork for Business Owners is delivered as a YouTube playlist rather than a Skilljar course. The frame shows the playlist with videos like "Claude Cowork for Dummies 101" (32:09, 107K views), "Claude Cowork 60 Minute MASTERCLASS" (1:01:12, 78K views), and "How to Use Claude Cowork 101" (30:32, 47K views). This is the practical, non-technical business track.

Courses 4-6: Coding and the API

Claude Code in Action is a practical walkthrough of using Claude Code to accelerate a development workflow. The page shows 15 lectures and the standard free enrollment with Share on X and Share on LinkedIn options. The narrator calls this one out as a highlight.

Claude Code for Business Owners is the second YouTube-playlist entry on the list, pairing the coding tool with a non-developer audience.

Building with the Claude API is where the track turns advanced. The course preview in the frame shows prompt-engineering diagrams: a "Message Profiling" panel and a token-selection probability tree (input tokens like "What / do / you / finish" mapping to a selection-probability column and a "New Token" output). It lists a "Prompt engineering" module with 16 lessons covering prompting strategies and testing approaches. The on-screen caption at this point reads "ADVANCED."

Courses 7-9: The Advanced Track (MCP, Skills, Subagents)

This is the part the video argues is worth chasing.

Intro to Model Context Protocol teaches connecting an agent to external tools. The learning-objectives frame lists: understand MCP architecture and the client-server communication model, build MCP servers that expose tools using the Python SDK, implement MCP clients, create resources for exposing data and prompts for pre-defined workflows, test and debug servers using the MCP Inspector, and choose between tools, resources, and prompts based on control patterns. Prerequisites include basic programming and the course targets engineers connecting Claude to external tools without boilerplate.

Intro to Agent Skills covers building, configuring, and sharing Skills in Claude Code. The description in the frame defines a Skill as reusable markdown instructions that Claude automatically applies to the right task at the right time, and says the course takes you from creating your first Skill to distributing them across teams and troubleshooting. A second frame for this course shows a data/privacy FAQ (questions about Skilljar, what learning data it collects, where data is stored, and deleting your learning account).

Intro to Subagents explains delegating tasks to isolated assistants that do work separately and return just the information you need, keeping the main context window clean. The curriculum frame lists: how sub-agents work (what happens when Claude Code spins up a separate context window, how inputs flow, how summaries come back), creating custom sub-agents with the /agents command (from code reviewers to documentation generators), designing effective sub-agents (structured input formats, obstacle reporting, limiting tool access), and when to use them plus common anti-patterns to avoid.

How to Get the List

The video ends with a call to action: to get the full list of links sent to you, follow the account and comment "university." A frame literally instructs viewers to "Take screenshot of list 📸" while displaying all nine titles with their URLs.

Key Takeaways

Anthropic's Claude learning path is nine free courses on anthropic.skilljar.com, with two of the business-owner courses delivered as YouTube playlists instead.
The track splits cleanly: courses 1-3 are foundations and business use, 4-6 are coding and the API, and 7-9 are the advanced agent-building track.
The advanced track is the differentiator -- MCP (connecting agents to tools via the Python SDK and MCP Inspector), Agent Skills (reusable markdown instructions auto-applied to tasks), and Subagents (delegating work to isolated context windows via /agents).
Several courses show concrete specs: AI Fluency is 14 lectures plus a certificate, Claude Code in Action is 15 lectures, and the API course's prompt-engineering module is 16 lessons.
Courses carry completion certificates and quizzes, so the path is structured learning, not loose tutorials.
To get the link list from the video itself, follow @sabrina_ramonov and comment "university," or screenshot the list frame directly.