mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Git workflow, debugging, autonomous TDD, and session-management skills.
Plugin dimmkirr-skills from marketplace dimmkirr in DimmKirr/skills. Commands read from the repo's manifest — 8 days ago.
git-commit-messageGenerate a structured conventional commit message for current changes and save it to .scratch/COMMIT.md. Use when the user asks to write, generate, or improve a commit message, is about to commit and needs a message, or another skill needs a commit message for staged changes. Do NOT use for PR descr
git-group-commit-messageGroup uncommitted changes by feature, write groups.json with files and commit messages, and generate a Taskfile that iterates and commits each group.
jetbrains-aicommitPrompt template for the JetBrains AI commit message feature (conventional commits). Not an agent workflow.
wdytGive a structured opinion on a topic - top 3 notable facts and top 3 probable theories.
ai-rebaseGenerate improved commit messages for a git commit range using parallel subagents, without creating commits.
cleanroom-analyzeCleanroom reverse engineering analysis (the "reader"). Examines vendor binaries and produces a hardware fact sheet using a legally grounded source hierarchy. Output is the ONLY artifact that crosses the barrier to the implementer. Must never produce code, pseudocode, or direct translations. Must nev
cleanroom-implementCleanroom implementation (the "writer"). Writes GPL-compatible driver code from a hardware fact sheet, public standards, and permitted open-source references ONLY. Must never access vendor binaries, disassembly, or the analyst transcript.
continue-debugAnalyze the output of the most recent debug run in .scratch/debug/ and iterate with ranked hypotheses.
data-engineerData quality verification skill for data engineering tasks. Use when working with data files, pipelines, or datasets that require validation. Triggers for: verifying file paths exist and are correct, checking file naming conventions, validating data ranges and types, auditing data quality, inspectin
devops-engineerUse when making infrastructure changes (nix, docker, MCP, terraform) that need validation, when an E2E test is too expensive for the iteration cycle, when you need to isolate a testable concern from a larger integrated system, or when the user asks to run/test something — present the cost/level pyra
drawio-diagramUse when working with .drawio diagrams — creating new architecture diagrams from scratch, cleaning up layout, analyzing positioning, fixing alignment or spacing issues, editing connectors, checking border overlaps, or any task involving mxGraphModel XML
frontend-qaFrontend QA workflow for Expo/React Native web apps. Use when asked to: test the app locally, verify all tabs/screens work, check if the OpenAPI schema is current, test API endpoints directly with curl/hurl, do visual QA with Playwright MCP screenshots, assess overall UI quality, or identify missing
git-group-commitGroup uncommitted changes into feature commits with conventional messages, and emit runnable commit artifacts. Use when the user asks to commit changes, write or improve a commit message, split work into multiple commits, or another skill needs commit messages for uncommitted changes. A single chang
git-pr-messageGenerate a pull request description from committed groups and save it to .scratch/PR.md.
git-releasePlan the next release from everything committed since the last tag — version, title, notes — and install a runner that creates it via tag push (CI creates the release) or GH draft (user creates the release), published with `git release --publish`.
implement-tdd-autonomouslyImplement the agreed work fully autonomously using strict TDD with judgment-call rules and fail-safes. Auto-detects the right verification mode based on context.
ios-qaiOS Visual QA using Go tests + Claude Vision. Runs Go test flows that capture per-step screenshots, then evaluates each screenshot against the app's design system (docs/DESIGN.md) and Linear user story acceptance criteria. Use when asked to: visually QA the iOS app, verify screens look correct, chec
kitcut-implementReproduce a working application from a kitcut-artifact/2.0 bundle (facts.json + checks.json + requirements/ + digests/; evidence/ optional and verification-only). Use when the user provides an artifact bundle directory and wants the described app rebuilt, reproduced, implemented, or incrementally re
pmProject manager for backlog grooming, ticket analysis, and implementation status tracking. Use for analyzing ticket relevance, checking implementation status against codebase, splitting partially-done tickets, managing labels/assignees, and maintaining backlog health.
react-native-blueprintExtract features, API contracts, and design system from React Native / Expo source code. Produces features.md, features.json, and design-system.md for re-implementation. Deterministic detection first, then LLM-enriched analysis.
reverse-wireframe-uiUse when creating wireframe diagrams from existing app screenshots or mockups. Triggers when user provides UI screenshots and wants drawio wireframes, component hierarchy documentation, or visual specs. Also use when reverse-engineering UI structure from images for design handoff, documentation, or
tickets-status-updatePost a structured status update comment (DONE / WORKING ON / BLOCKED ON) to the current ticket in Linear/Jira.
vector-designerUse when creating logos, icons, SVG graphics, or any vector design work requiring visual iteration. Use when user asks to design, draw, animate, or create visual assets like logos, favicons, badges, or diagrams in SVG/PNG format. Use when adding CSS-based logo animations or brand reveal effects.
visual-qaVisual QA for any UI via screenshot review. Structure-down checklist in four layers: general rules (layout, hierarchy, Gestalt grouping, spacing, typography, color, accessibility, modals, states, consistency) for all UIs; web/desktop rules (responsive, stacking, fonts, scroll, focus); mobile rules (
cohesion-reviewCohesion analysis of a pipeline, workflow, skill, data layout or module. Use when the user asks why a workflow redoes work whose inputs did not change, which steps should be cached or skipped, whether something belongs in rows or fields, who owns a file with two writers, whether expensive steps amor
setup-appstore-releaseterminal-image-viewerShow an image inline in the user's terminal from inside a devcell container — screenshots, renders, diffs, generated art. Use when the user asks to see an image, or when reviewing something visual (illustration edits, page renders, chart output, UI screenshots) would be faster to judge by eye than b
decomposition-reconstructionUse when dealing with complex systems that need simplification, identifying bottlenecks or critical failure points, redesigning architecture or processes for better performance, breaking down problems that feel overwhelming, analyzing dependencies to understand ripple effects, user mentions "this is
linear-workerUse when the user wants to monitor Linear tickets assigned to a user and react to comments, status changes, or stale tickets. Use when setting up a recurring routine or loop that watches Linear for activity and takes action (responds, flags, nudges, summarizes).
loadLoad .scratch/CONTINUE.md, reconcile it with the workspace, and continue the saved work.
markdown-to-pdfConvert markdown to PDF, create PDF from .md file, render markdown as print-ready document. Use when: "convert markdown to pdf", "create pdf out of this file", "render this as pdf", "make a pdf from markdown", or any request to turn a .md file into a styled PDF document.
saveSave current session state to .scratch/CONTINUE.md and prepare for a safe user-controlled exit.
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
8 skills · 1 MCP server
Perform AI code review on Git diffs — supports workspace changes, branch ranges, and single commits with concurrent per-file analysis, codebase search, and deep context-aware review.
2 commands
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams
6 skills · 7 commands · 4 agents
Long-term semantic memory for Claude Code, powered by OpenViking
5 skills · 1 command · 1 MCP server · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.