← Back to research
·3 min read·company

Anthropic Skills

Anthropic's official Agent Skills catalog — a public repository of SKILL.md-based instruction packages that teach Claude how to complete specialized tasks via progressive disclosure.

Key takeaways

  • Skills are folders with a SKILL.md file that Claude loads dynamically — progressive disclosure means only relevant context enters the window.
  • Works across Claude Code (as plugins), Claude.ai (built-in for paid plans), and the Claude API — full platform coverage.
  • The repo includes production document skills (docx, pdf, pptx, xlsx) that power Claude's native document capabilities, plus open-source examples.

FAQ

What are Anthropic Agent Skills?

Agent Skills are self-contained folders of instructions, scripts, and resources that Claude loads on-demand to improve performance on specialized tasks. Each skill has a SKILL.md file with YAML frontmatter (name + description) and markdown instructions.

How does progressive disclosure work in Agent Skills?

Claude first reads only the skill's name and description (from frontmatter). If the skill is relevant, it reads the SKILL.md body. Only if needed does it load additional files (scripts, references, assets). This keeps the context window lean.

How do you install skills in Claude Code?

Run `/plugin marketplace add anthropics/skills` to register the marketplace, then `/plugin install document-skills@anthropic-agent-skills` or `/plugin install example-skills@anthropic-agent-skills` to install skill sets.

What types of skills are in the catalog?

Creative & Design (art, music, design), Development & Technical (web app testing, MCP server generation), Enterprise & Communication (branding, comms), and Document Skills (docx, pdf, pptx, xlsx creation/editing).

Executive Summary

anthropics/skills is Anthropic's official public repository of Agent Skills — self-contained instruction packages that teach Claude how to perform specialized tasks. With ~73k GitHub stars, it's both a reference implementation and a usable catalog spanning creative, technical, enterprise, and document workflows. The key innovation is progressive disclosure: skills load context incrementally so Claude only reads what it needs, keeping the context window efficient.

AttributeValue
MaintainerAnthropic
Stars~73k
LicenseApache 2.0 (examples); source-available (document skills)
Specagentskills.io
PlatformsClaude Code, Claude.ai, Claude API

The SKILL.md Format

Every skill is a folder with a required SKILL.md file:

---
name: my-skill-name
description: A clear description of what this skill does and when to use it
---

# My Skill Name

[Instructions Claude follows when this skill is active]

## Examples
- Example usage 1

## Guidelines
- Guideline 1

Optional subdirectories extend the skill:

  • scripts/ — Executable code (Python, Bash, etc.)
  • references/ — Documentation loaded as needed
  • assets/ — Templates, fonts, icons used in output

Progressive Disclosure

This is the core design principle. Rather than dumping all context upfront:

  1. Level 1 — Frontmatter only: Claude sees skill name + description (a few tokens)
  2. Level 2 — SKILL.md body: If relevant, Claude reads the full instructions
  3. Level 3 — Supporting files: Only loaded when the task requires them

This is why skills scale better than stuffing everything into a system prompt. An agent can have dozens of skills registered but only pay the token cost for the ones actually used.

Key Skill Categories

Document Skills (Source-Available)

The production skills that power Claude's native document capabilities:

  • docx — Word document creation and editing
  • pdf — PDF extraction and generation
  • pptx — PowerPoint creation
  • xlsx — Excel spreadsheet generation

Example Skills (Apache 2.0)

Open-source demonstrations across domains:

  • Creative & Design — Art, music, design workflows
  • Development & Technical — Web app testing, MCP server generation
  • Enterprise & Communication — Branding, communications, workflows

Platform Support

PlatformHow to Use
Claude Code/plugin marketplace add anthropics/skills then install specific skill sets
Claude.aiBuilt-in for paid plans; custom skills can be uploaded
Claude APIUpload via Skills API endpoint

Partner Ecosystem

Anthropic highlights third-party skills, starting with Notion's official skill set. The agentskills.io specification aims to make skills a cross-platform standard, not just a Claude feature.

Analysis

Agent Skills formalize something the agentic engineering community has been doing informally — giving agents structured instructions in files. What Anthropic adds is:

  1. A standard format (SKILL.md with frontmatter) that's simple enough to adopt
  2. Progressive disclosure as a first-class concept, not an afterthought
  3. Platform integration across CLI, web, and API surfaces
  4. A marketplace model via Claude Code plugins

The 73k stars suggest strong community interest. The real question is whether agentskills.io gains cross-vendor adoption or remains Claude-specific. For now, the format is simple enough that any agent framework could adopt it — it's just markdown files in folders.