mattpocock/skills

Skills for Real Engineers. Straight from my .agents directory.

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 41 minutes ago
Added to GitGenius on April 12th, 2026
Created on February 3rd, 2026
Open Issues & Pull Requests: 321 (+0)
Number of forks: 18,463
Total Stargazers: 213,916 (+57)
Total Subscribers: 1,233 (-1)

Issue Activity (beta)

Open issues: 315
New in 7 days: 61
Closed in 7 days: 16
Avg open age: 13 days
Stale 30+ days: 112
Stale 90+ days: 1

Recent activity

Opened in 7 days: 55
Closed in 7 days: 15
Comments in 7 days: 62
Events in 7 days: 94

Top labels

  • enhancement (10)
  • ready-for-human (5)
  • wontfix (5)
  • needs-info (3)
  • needs-triage (3)
  • duplicate (2)
  • documentation (1)
  • ready-for-agent (1)

Repository Insights (GitGenius)

Median issue/PR response: 5.4 hours
Mean response time: 3.6 days
90th percentile: 11.1 days
Tracked items: 497

Most active contributors

Sign in to see contributor activity.

How this project is maintained

Around half of the issues opened in the past year never receive a reply. 100% of open issues come from outside the core team, so the backlog reflects real-world use rather than internal planning. Almost all tracked open issues have seen activity in the last three months. Only 10% of issues opened in the past year have been closed. Three people close 95% of everything that gets resolved.

Detailed Description

The mattpocock/skills repository is a collection of agent skills designed to improve how AI coding assistants like Claude work on real engineering projects. Written primarily in Shell, the repository contains practical workflows and prompts that address common failure modes when using AI for software development. The repository describes itself as containing skills "straight from my .claude directory," representing techniques that creator Matt Pocock uses daily in actual engineering work rather than theoretical approaches.

The most frequently applied issue labels are enhancement, ready-for-human, and wontfix, suggesting an organized approach to issue management and feature development.

The core philosophy behind these skills addresses four major problems in AI-assisted development. The first is misalignment between what developers want and what agents produce. The repository includes skills like /grill-me and /grill-with-docs that prompt agents to ask detailed clarifying questions before beginning work, establishing shared understanding before implementation starts. The second problem is agent verbosity, which the repository solves through building a shared domain language documented in a CONTEXT.md file. This approach reduces the jargon gap between agents and projects, making codebases easier to navigate and reducing token consumption. The third problem is code that doesn't work, addressed through skills like /tdd that implement red-green-refactor loops and /diagnosing-bugs for structured debugging practices. The fourth problem is architectural degradation, where codebases become complex and difficult to change, tackled through /improve-codebase-architecture and /to-prd skills that encourage design thinking throughout development.

The repository organizes skills into two categories: user-invoked skills that developers explicitly call to orchestrate workflows, and model-invoked skills that agents can reach for automatically when tasks fit their purpose. Engineering skills include /ask-matt as a router over other skills, /grill-with-docs for requirements gathering with domain modeling, /triage for issue state management, /to-issues for breaking specifications into vertical slices, and /setup-matt-pocock-skills for initial configuration. The quickstart process involves running a skills.sh installer, selecting desired skills, and configuring issue tracker preferences and triage labels.

The repository is classified across multiple domains including Programming, Web Development, TypeScript, React, JavaScript, Learning Resources, Code Examples, Software Engineering, and Technical Education. The skills are designed to be small, composable, and adaptable to different coding agents and workflows, emphasizing that they work with any model rather than being locked to specific platforms.

skills
by
mattpocockmattpocock/skills

Repository Details

Fetching additional details & charts...