Interviews you relentlessly to sharpen a plan or design before you commit to it.
Runs only when you type /grill-me
Gives coding agents ready-made workflows for planning, reviewing, testing, writing and shipping software.
Interviews you relentlessly to sharpen a plan or design before you commit to it.
Runs only when you type /grill-me
Interviews you hard about a plan while also writing decision records and a glossary as you go.
Runs only when you type /grill-with-docs
Scans a codebase for structural improvement opportunities, shows them in an HTML report, then digs into your pick.
Runs only when you type /improve-codebase-architecture
Builds features and fixes bugs test-first, writing failing tests before code and refactoring after.
Compacts the current conversation into a handoff document another agent can pick up.
Runs only when you type /handoff
Configures a repo's issue tracker, triage labels and doc layout before first use of the other skills.
Runs only when you type /setup-matt-pocock-skills
Builds a throwaway prototype to test whether a design or state model feels right.
Moves issues and outside pull requests through triage steps, sorting, verifying and writing briefs for agents.
Runs only when you type /triage
Builds and sharpens a project's domain model, glossary and architecture decision records.
Helps design module interfaces and find places where code can be split into deeper, simpler modules.
Teaches you a new skill or concept step by step inside your current workspace.
Runs only when you type /teach
Runs a structured diagnosis loop for hard bugs and performance regressions until the cause is found.
Reviews changes since a commit or branch against your coding standards and the original spec, side by side.
Implements a piece of work based on a spec or a set of tickets.
Runs only when you type /implement
Recommends which skill or workflow in this collection fits your current situation.
Runs only when you type /ask-matt
Maps a large piece of work as decision tickets on your tracker and resolves them one at a time.
Runs only when you type /wayfinder
Turns your current conversation into a written spec and publishes it to the project issue tracker.
Runs only when you type /to-spec
Breaks a plan or spec into small tickets with blocking links and publishes them to your tracker.
Runs only when you type /to-tickets
Researches a question using trusted primary sources and saves the findings as a Markdown file.
Generates an interactive bash wizard that walks you through steps only a human can do.
Turns a decision you can't fully answer into a questionnaire someone else can fill in.
Runs only when you type /to-questionnaire
Writes and edits documents meant for agents, such as skills, AGENTS.md and CLAUDE.md files.
Sets up Claude Code hooks that block dangerous git commands like push, reset --hard and clean.
Sets up Husky pre-commit hooks with lint-staged, Prettier, type checking and tests.
Assembles raw material into a journey of beats, explaining each term before a beat relies on it.
Runs only when you type /writing-beats
Shapes raw material into an article, working paragraph by paragraph.
Runs only when you type /writing-shape
Mines raw fragments and notes from your material before any structure is imposed.
Runs only when you type /writing-fragments
Creates exercise folders with sections, problems, solutions and explainers that pass linting.
Stops and re-explains the last message when it didn't land clearly.
Runs only when you type /wait-what
Rewrites test files to replace `as` type assertions with @total-typescript/shoehorn helpers.
Interviews you about the specs for workflows you want to build in this workspace.
Runs only when you type /loop-me
Hands the current conversation to a fresh background agent that continues the work right away.
Runs only when you type /claude-handoff
Wires dependency-cruiser into a TypeScript repo so each package is only reachable through its entry points.
Runs only when you type /setup-ts-deep-modules
Turns a written spec and its tickets into working code.
Runs only when you type /implement-spec
Runs a retrospective on a coding session to capture what went well and what to change.
Runs only when you type /retro
Writes the body text for a pull request describing your changes.
Pursues a long-running goal in one session by coordinating helper agents on your behalf.
Runs only when you type /chief-of-staff
Sets up a shared vocabulary for the tasks and services your automated software pipeline plugs into.
Runs only when you type /software-factory
Stress-tests your plan, decision or idea with relentless questions to expose weak spots.