Agent skills
Skills for AI coding agents, each with what it costs to load, how much of it is scripts and what it touches on your machine.
How to read a skill
What it costs to keep
Idle in context, in use when it runs, and up to with its linked files.
Scripts or instructions
Script-led, mixed or model-led: how much the model has to work out itself.
What it can reach
Shell, network, file writes, git and credentials, declared or detected.
Quality, security, repo health
Out of 100, each with what it is made of. Turn a card over to see them.
Ask Matt
328,754A router over Matt Pocock's engineering skills that tells you which skill or flow fits your situation, from idea to shipped code.
Code Review
328,754Developer Tools
Reviews the changes since a fixed point along two axes, repo standards and spec match, with parallel sub-agents and a Fowler smell baseline.
Codebase Design
328,754Developer Tools
Shared vocabulary for deep modules: a lot of behaviour behind a small interface, placed at a clean seam and testable through that interface.
Domain Modeling
328,754Builds and sharpens a project's domain model by challenging terms, stress-testing scenarios and updating GLOSSARY.md and ADRs inline.
Grill With Docs
328,754A user-invoked skill that runs the grilling interview and the domain-modeling discipline together, leaving a glossary and ADRs behind.
Implement
328,754Developer Tools
Builds the work in a spec or ticket by driving TDD at pre-agreed seams, then runs code-review and commits to the current branch.
Implement Spec
328,754Developer Tools
Implements a whole spec on one integration branch by running implementer subagents in parallel across the ready tickets.
Improve Codebase Architecture
328,754Developer Tools
Scans a codebase for deepening opportunities, shows them in a visual HTML report, then interviews you through the candidate you choose.
PR
328,754Developer Tools
A pull request body template with a summary as the smallest useful visual, before and after evidence, and a merge-danger call.
academy-guide
197,764AI Fluency
When someone asks how to use Claude or a Claude product, Claude answers first, then points to one or two strongly matching Claude Academy courses, tutorials or use cases.
mcp-builder
197,764Developer Tools
Anthropic's skill for building MCP servers: research the API, implement tools in TypeScript or Python, test with the Inspector, then measure the server with an LLM evaluation harness.
theme-factory
197,764Design And Visual Art
Anthropic's theming skill: ten preset color palettes and font pairings Claude applies to slides, docs, reports and HTML pages, or a custom theme made on the spot.
algorithmic-art
197,763Design And Visual Art
Anthropic's generative art skill: Claude writes an algorithmic philosophy, then builds it as a seeded p5.js sketch in an interactive HTML viewer.
brand-guidelines
197,763Design And Visual Art
Anthropic's brand skill: the official colors, Poppins and Lora typography and accent rules Claude applies to make any artifact look on-brand.
canvas-design
197,763Design And Visual Art
Anthropic's skill for posters and static art: Claude writes a design philosophy as an art movement, then expresses it as one crafted PDF or PNG canvas.
discernment-nudge
197,763AI Fluency
After a substantive answer the user may act on, Claude appends two or three specific follow-up questions that help check facts, probe reasoning and notice missing context.
doc-coauthoring
197,763Writing And Communication
A three-stage workflow for writing docs with Claude: gather context, build each section through brainstorming and edits, then test it on a fresh Claude reader.
frontend-design
197,763Design And Visual Art · Developer Tools
Design guidance that pushes Claude toward distinctive, intentional UI instead of templated defaults, from palette and type to layout and copy.
internal-comms
197,763Writing And Communication
Anthropic's skill for writing internal communications: 3P updates, company newsletters, FAQ roundups and other company comms, each in a fixed house format.
humanizer
71,585Writing And Communication
Rewrites AI-sounding text so it reads like a person wrote it: it marks 26 known tells, drafts a rewrite and checks that every claim survives.