A marketplace of Claude skills — packaged instruction files you install instead of writing the prompt yourself.

Pick this when

Someone has already solved the instruction problem you're about to solve — sort by usage count and take theirs.

What makes it different

Usage counts as the quality signal — the top skill is a 1.4M-download skill-finder

Skills convert cleanly into system prompts for other tools, which is how BC3/BC4 used them inside n8n

Covers surprisingly specific ground (Azure Kubernetes advisor, web scraping, content humanizer)

What it does

Distributes reusable skill files for Claude

Ranks them by downloads so picking is fast

Supplies instruction text that can be lifted into an n8n agent's system prompt

What it does not do

Common wrong expectations

Doesn't guarantee fit — a skill is someone else's opinion about your task

Doesn't stay put: a skill pasted into another tool stops receiving its author's updates

Where it sits in a stack

The instruction layer. Where a meta prompt is one you build, a skill is one you borrow.

Shelf

Marketplaces & directories · Productivity & source apps

Our status · watch

The pipeline's instructions are bespoke, but the borrow-then-adapt pattern is directly usable.

For your projects

The skill→system-prompt conversion is the transferable trick: the extraction protocol could ship as a skill rather than a document that has to be re-read every sitting.

Freshness

current

Related tools

builds onClaudecomplementsTaste skill / content humanizercomplementsn8n

claude: skills are a Claude feature

taste-skill: one of the skills used live in BC4

n8n: skill text becomes an agent node's system prompt

Every moment the course touched it (15)

SessionCoverageMomentWhat happened
Basecamp 3: Introduction to n8n — Idemonstrated1:16:37Skill marketplace used twice live (Azure Kubernetes advisor, web scraping); pick by usage count
Basecamp 4: Introduction to n8n — IIdemonstrated1:27:55Content humanizer skill (based on Claude's official humanizer) and the taste skill; top skill on the site is a 1.4M-download skill-finder skill
Session 3: Building Elegant Landing Pagesdemonstrated0:30:04Marketplace toured: Anthropic frontend-design skill, web-design guidelines, Remotion practices; pick by install count (an SEO-audit skill at 123k installs shown)
Session 7: The Outreach Enginedemonstrated2:26:35Skill marketplace (698k+ skills); source of the Humanizer skill (24k+ stars) composed into the email-copy skill
Session 9: Introduction to Cursor & its principlesdemonstrated1:02:14Skill marketplace: superpowers (his universal core; a plugin technically) and extract-design-system (124k installs) both installed by prompt
Session 10: Building on Cursor & connect with AI workflowsdemonstrated1:19:10Source of the security-review skill; superpowers reused from Session 9 for the PRD interview
Session 12: Claude Code — Design & Building Agents (App-First: a Game, a GTM Board, and a Live Deploy)demonstrated1:22:44Marketplace skill installed by pasted prompt as the install demo
AI Sprint: Claude Code with Antigravity — Day 2 (CLAUDE.md, the 50% Context Rule, and the Sketch-to-App Build)demonstratedThe interface-quality layer (~65k stars as-heard), installed live
AI Sprint: Claude Code with Antigravity — Day 2 (CLAUDE.md, the 50% Context Rule, and the Sketch-to-App Build)demonstratedMarketplace picks by installs/stars: e2e-testing patterns, security best-practices; Anthropic's own web-app-testing spotted
Session 14: Claude Code — Advanced Workflowsexplained1:08:17The incumbent being disrupted: 836k skills (as heard), Vercel-run, no payments or real vetting
Session 14: Claude Code — Advanced Workflowsexplained1:23:29Prescribed pre-install audit for any third-party skill; link dropped in chat
Session 4: How to Build Micro Prototypesmentioned0:55:02Recommended for the landing page in front of the app (GummySearch's landing/product split as the model)
Session 8: The Conversion Playbookmentioned1:47:10Source of the taste skill ('anti AI slop frontend') recommended for de-slopping generated pages, and the Humanizer reused throughout
Session 11: Overview of Claude Code & Key Concepts (Installs, Modes, Context Files, Skills, Hooks & MCPs)mentioned1:38:27'Not exactly the same thing' as Claude skills; marketplace trust rule applied in Q&A
Session 12: Claude Code — Design & Building Agents (App-First: a Game, a GTM Board, and a Live Deploy)mentioned2:38:38Identified in Q&A as an agent-skills directory like the marketplace demoed

← All tools by category