3 Claude Code Commands Every Writer Should Know
A 10-minute screen-share walkthrough of the three Claude Code slash commands that actually matter for writers -- live demo, real file tree, no developer jargon.
December 26th 2025A 20-minute walkthrough of Anthropic's official Claude Code skills repository — install 16 packaged skills in two minutes, then combine them with context profiles to build a co-writing system that actually knows you.
Most people using Claude Code are leaving half the toolkit on the table. Alex McFarland opens with the kind of claim that stops the scroll — ninety-nine percent don't know this exists — then spends twenty minutes proving it by installing Anthropic's official skills repository live and watching it generate a branded nine-slide presentation in under five minutes.
stated at 00:04“I'm gonna show you how to upgrade your co writer inside of Claude by using Anthropics official repository of Claude's skills.”delivered at 08:00

States the promise ('upgrade your co-writer'), acknowledges the Claude Code Masterclass as prerequisite, sets up the skills concept.

Four-property definition: packaged instructions, teach Claude HOW, write once/use forever, auto-activate. Not traditional prompts.

Two-column diagram — Basic (.claude/skills/name/SKILL.md) vs With Resources (adds templates/ and references/). ICP context profile as the worked example.

Flow diagram: ask → scan available skills → find match → load skill → execute automatically. Notes near-100% hit rate in practice.

Live terminal demo: type `claude` → /plugin → Marketplaces tab → Add Marketplace → paste anthropics/skills → enter. Error shown confirms it was already installed. Two categories: document skills and example skills.

Queries Claude for available skills. Full list: PDF, spreadsheet, document, presentations, frontend-design, canvas-design, algorithmic-arts, Theme Factory, MCP Builder, Skill Creator, web app testing. Then 2x2 grid: frontend-design (TOP PICK), pdf (TOP PICK), xlsx (ESSENTIAL), skill-creator (ESSENTIAL).

Core message: you don't invoke manually. Describe what you want, Claude scans skills and picks. Caveat: watch the thinking stream to confirm activation; cancel + restart if it skips.
Standalone skills are good. Combined with voice DNA, ICP JSON, business profile, brand colors folder — that's supercharging. The co-writer becomes context-aware.

Opens Claude Code, references Substack newsletter, sends 'create a presentation using your front end design skill.' Watches skill activate in thinking stream. Output: 9-slide HTML presentation with dark editorial aesthetic, glowing effects, code callouts.

Shows raw output, admits it doesn't look perfect yet. Explains the iteration loop: point Claude to brand folder, images, guidelines — keep rebuilding. His own polished presentations came from this loop.
Same newsletter, 'create a PDF I can use as a lead magnet.' PDF skill writes a Python script internally, generates PDF, saves to project root. Same brand iteration caveat applies.

CTA: Substack, waitlist for co-writer system, comment what you built. Teases next: deep dives, chaining skills together, full Claude features breakdown for writers.
Four-property definition that positions skills as a different category from prompts or procedures.
Two-tier architecture for skills — markdown-only vs markdown + supporting files. The ICP questionnaire is the worked example of the With Resources tier.
The layering of context profiles that makes skills output match your brand instead of generic AI output.
“99% of people don't even realize that they can be doing this.”
“You don't invoke them manually. Just describe what you want, and Claude figures out the skill.”
“If you have standalone skills, that's great. They work incredible. But when you combine these with our cowriting system, with our context profiles — that's when you're really supercharging your cowriter.”
“Write once and use forever and then auto activate.”
“Make sure to check out my Substack. Subscribe if you want any of the assets I use in my videos. Sign up for a wait list if you're interested in getting my entire collection of skills and system prompts.”
Soft close, no hard sell. Three parallel CTAs: subscribe, waitlist, comment. Waitlist is the money ask buried in the middle.
McFarland built his presentation with the exact frontend-design skill he's demonstrating — the deck IS the proof of concept, and that closed loop is what makes the tutorial land.
00:00
00:22
00:37
00:52
01:07
01:21
01:36
01:51
02:06
02:21
02:36
02:51
03:06
03:21
03:36
03:50
04:05
04:20
04:35
04:50
05:05
05:17
05:35
05:50
06:05
06:19
06:29
06:49
07:07
07:18
07:34
07:49
08:04
08:19
08:34
08:48
09:03
09:18
09:33
09:48
10:03
10:18
10:33
10:48
11:03
11:17
11:32
11:47
12:02
12:17
12:32
12:47
13:02
13:17
13:32
13:46
14:01
14:16
14:31
14:46
15:01
15:16
15:31
15:46
16:01
16:15
16:30
16:51
17:00
17:15
17:30
17:45
17:58
18:15
18:30
18:44
18:59
19:18
19:29
19:44A 10-minute screen-share walkthrough of the three Claude Code slash commands that actually matter for writers -- live demo, real file tree, no developer jargon.
December 26th 2025A 26-minute walkthrough of Anthropic's new Cowork tab — three real use cases for writers who are not developers.
January 14th 2026A 37-minute tutorial showing non-developers how to replace one-off prompts with a persistent, file-based AI co-writing system built inside Claude Code.
December 10th 2025A 15-minute walkthrough of how Claude Code gets eyes: setup, tab groups, permissions, and three real browser automation use cases with an honest speed warning.
December 19th 2025Theo goes all-in on Claude Code over the holiday break — six parallel instances, no IDE opened, two projects from scratch — and comes back with a changed worldview on writing code.
January 6th 2026A 17-minute framework for when running multiple Claude Code terminals actually makes you faster — and when it just burns money.
January 19th 2026