smll-ai

smll-ai / mattpocock-skills

Public

Skills for Real Engineers. Straight from my .claude directory. (github/mattpocock)

16
4
100% credibility
Found May 02, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A curated set of prompts and guides that empower AI coding assistants to follow engineering best practices for planning, testing, debugging, and improving real software projects.

How It Works

1
🔍 Discover Helpful Skills

You hear about a collection of smart tricks that make AI coding buddies super effective for building real apps, shared by an experienced engineer.

2
📦 Grab the Skills Pack

Follow the simple guide to add these tricks to your AI helper's toolbox in just seconds.

3
⚙️ Set Up for Your Project

Chat with your AI helper to connect it to your task list and project notes, making everything ready to go.

4
💬 Grill Your Ideas

Before starting work, let the AI ask tough questions to clarify exactly what you want to build, avoiding misunderstandings.

5
🧪 Test and Debug Smartly

Use guided loops to write tests first, hunt down bugs step-by-step, and keep your code running smoothly.

6
🏗️ Polish Your Project

Review and refine the overall design regularly, turning messy code into clean, easy-to-change structure.

🚀 Build Amazing Apps

Now your AI helper creates reliable software faster, helping you ship your best work ever.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mattpocock-skills?

Matt Pocock's skills GitHub repo delivers a pack of slash commands for AI coding agents like Claude Code, Copilot CLI, and Copilot VSCode. It tackles real engineering pain points—misaligned plans, verbose outputs, buggy code, and architectural drift—by providing grilling sessions, TDD loops, bug diagnosis, and codebase refactoring tools. Install via a 30-second npx command, pick your agent and issue tracker (GitHub, Linear, or local), and run setup to get commands like /grill-with-docs or /tdd working instantly.

Why is it gaining traction?

Unlike bloated agent frameworks that lock you into rigid processes, these skills are small, model-agnostic, and composable, drawing from decades of pragmatic engineering to fix AI-specific failures like poor feedback loops. Developers hook on the shared language docs that cut verbosity by defining project jargon upfront, plus red-green-refactor TDD that actually produces working code. Skills GitHub Claude and Copilot users notice tighter alignment and faster debugging without losing control.

Who should use this?

Fullstack engineers using skills GitHub Anthropic or OpenClaw for production apps, especially those triaging GitHub issues or debugging performance regressions. Teams building with Copilot in VSCode who hit "the agent didn't get it" walls, or solo devs wanting /diagnose loops for hard bugs. Skip if you're just prototyping—it's for shipping maintainable code.

Verdict

Grab it if you're deep into Claude Code or Copilot workflows; the docs are solid and battle-tested by Matt Pocock. With 16 stars and 1.0% credibility score, it's early-stage and unproven at scale, but low risk to try on your next repo.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.