Week 12 — Readings & Resources · Claude Cowork II: Skills, Connectors & Artifacts

Course: Using Artificial Intelligence (AI 101) · Silver Oak University (fictional sample) · Prof. Quinn
Objective covered: Objective 5 — Use Claude Cowork to build agentic workflows — skills, connectors (MCP), live artifacts, and plugins.

How to use this page

Everything here is a link to an external resource. Nothing needs to be downloaded. This week, the most important resources are the official Anthropic documentation — these are the only reliable source for what Cowork features actually do, since product details change quickly.

A reminder from the course's core discipline: before trusting any claim about an AI feature — from a video, a blog post, or even an AI assistant — verify it against the official docs. This week that habit is load-bearing: the quiz and Studio both require you to know what's accurate.

Total time: roughly 40–50 minutes if you do all four primary readings; far less if you skim. Do the Anthropic docs first — they're short and definitive.

① The Official Anthropic Docs — Read These First (load-bearing)

Maps to Lecture Segments 2–6. These are the verified sources for every Cowork feature taught this week. Verified live before this module was published.

Use connectors to extend Claude's capabilities (Anthropic)
🔗 https://support.claude.com/en/articles/11176164-use-connectors-to-extend-claude-s-capabilities
Why it's first: the primary official doc on connectors — what they are, how to add one, what permissions mean, and what the connectors directory looks like. Read this before anything else this week.
⏱ ~8 min

Anthropic Connectors Directory FAQ (Anthropic)
🔗 https://support.claude.com/en/articles/11596036-anthropic-connectors-directory-faq
Why it's here: answers the most common questions about the directory — how connectors are vetted, what permissions mean, and how the directory is maintained.
⏱ ~5 min

Use plugins in Claude (Anthropic)
🔗 https://support.claude.com/en/articles/13837440-use-plugins-in-claude
Why it's here: the official doc explaining what a plugin is (a bundle of skills + connectors + sub-agents), how to find and install plugins, and how the Customize menu works.
⏱ ~5 min

Extend Claude with skills (Claude Code Docs — Anthropic)
🔗 https://code.claude.com/docs/en/skills
Why it's here: explains what a skill is, how SKILL.md files work (YAML frontmatter + Markdown instructions), and the difference between built-in skills and custom skills.
⏱ ~8 min

Use live artifacts in Claude Cowork (Anthropic)
🔗 https://support.claude.com/en/articles/14729249-use-live-artifacts-in-claude-cowork
Why it's here: confirms how live artifacts work — they refresh with current data from your connected connectors, live in their own tab, and keep version history.
⏱ ~5 min

② What MCP Is — The Open Standard Behind the Connectors

Maps to Lecture Segment 3. MCP is the open standard created by Anthropic that makes connectors possible — it's not just a Claude-specific feature.

Getting started with custom connectors using remote MCP (Anthropic)
🔗 https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp
Why it's here: explains MCP (Model Context Protocol) at the level needed for this course — what it is, why Anthropic created it as an open standard, and how custom connectors work for people who want to go deeper.
⏱ ~8 min

MCP official site (Anthropic)
🔗 https://modelcontextprotocol.io
Why it's here: the project homepage for MCP — confirms it's an open standard, not proprietary to Claude alone. Useful for students who want to see the spec or understand the ecosystem.
⏱ ~5 min (skim)

③ Background: Why These Features Matter

Optional background — skip if you're short on time; the official docs above cover everything tested this week.

Claude.ai official product page (Anthropic)
🔗 https://claude.com
Why it's here: the canonical starting point for Claude — if you want to see how Anthropic describes Cowork features publicly, this is the place. No specific claims are made here about features not covered in the docs above.
⏱ 5 min (browse)

Release notes — Claude (Anthropic)
🔗 https://support.claude.com/en/articles/12138966-release-notes
Why it's here: shows how quickly Cowork features have evolved — a reminder to always verify current behavior against the docs rather than assuming last month's notes are still accurate.
⏱ ~5 min (skim)

④ Get Set Up for the Studio

The Studio this week asks you to either (a) use a built-in Cowork skill to generate a document, or (b) connect a low-risk connector, or (c) build a live artifact. You'll need Claude Cowork desktop for options (a) and (c).

• Claude Cowork / Claude desktop app (Anthropic) 🔗 https://claude.ai/download (if not yet installed)
• Get started with Claude Cowork (Anthropic) 🔗 https://support.claude.com/en/articles/13345190-get-started-with-claude-cowork
• Browse skills, connectors, and plugins in one directory (Anthropic) 🔗 https://support.claude.com/en/articles/14328846-browse-skills-connectors-and-plugins-in-one-directory

(If you cannot install Claude Cowork, the Studio provides a "design and document" alternative path — read it carefully.)

Pick-one quick path (≈21 min total)

Short on time? Do exactly these three and you'll be ready for the quiz:
1. "Use connectors to extend Claude's capabilities" (group ①) — ~8 min.
2. "Use plugins in Claude" (group ①) — ~5 min.
3. "Extend Claude with skills" (group ①) — ~8 min.

Heads-up (links rot): these point to outside sites that occasionally move pages. If a link ever fails, go to support.claude.com and search the topic. For MCP, go to modelcontextprotocol.io. All resources are links only — nothing is downloaded or redistributed from these sources.