mattpocock/skills is the most-watched community skills directory in the Claude Code ecosystem. Matt Pocock — a TypeScript educator and longtime tooling builder — published his personal `.claude/skills/` folder as a public repo, treating it as a reference set for what production agent skills look like in practice. The repo crossed 48,564 GitHub stars in late April 2026 after a +6,175 star day, and held the #2 spot on GitHub trending for six consecutive days. Each skill is a single SKILL.md file with YAML frontmatter (name, description) plus referenced scripts where needed; the collection emphasizes small, sharp, single-purpose primitives — TDD workflows, GitHub triage, breaking architecture into vertical slices, and the engineering motions Pocock uses every day. The strategic insight the repo demonstrates is portable: break engineering experience into small skills, then let the agent combine them by scenario. The personal-`.claude/`-as-public-asset format Pocock pioneered has become the durable pattern of the 2026 skills directory cycle, validated by Anthropic shipping the SKILL.md spec as an open standard in December 2025.
mattpocock/skills is what you get when an opinionated TypeScript educator publishes the .claude/skills/ folder he actually uses every day. The repo is a practical reference — small, sharp, single-purpose primitives that demonstrate what production agent skills look like in 2026.
This is not a comprehensive skills catalog (see obra/superpowers for that) and not a directory of community-contributed skills (see ComposioHQ/awesome-codex-skills). It is one engineer’s working set, made public, and that constraint is the point: the skills are battle-tested in Pocock’s own workflow before they ship.
Three reasons mattpocock/skills is the most-watched skills repo on GitHub trending:
~/.claude/skills/, and have a working set on day one.Pairs naturally with obra/superpowers, warp, and cc-switch for a complete agent-skills runtime stack.
AWS's AI-powered coding assistant that helps developers build, deploy, and optimize applications on AWS with code generation and transformation.
Open-source AI pair programming tool that works in your terminal to edit code across your entire repository.
Open-source AI coding harness builder that makes AI coding workflows deterministic and repeatable via YAML-defined DAG workflows.