mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Build a product with Claude as lead and parallel agents: waves, briefs, quality gate, releases, languages, What's new, polish, security and accessibility.
Plugin build-framework from marketplace pruiter-plugins in peterruiter/claude-build-framework. Commands read from the repo's manifest — 26 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
a11y-passFull accessibility pass to WCAG 2.2 AA - axe-core through Playwright over every main page in light and dark at 1440/390 (and 320 for reflow), probes for skip link, one h1, lang, visible and unobscured focus, 24px targets and sideways scrolling, a keyboard script for dialogs/drawers/tabs, manual chec
agent-briefWrite one complete agent brief from the framework's template for a single piece of work (feature, fix, polish, translation, accessibility or security item), with context, scope, reserved migration slot, scratch database, ports, file ownership, the standing rules (all configured languages, WCAG 2.2 A
check-toolsCheck that every tool a profile needs is installed and recent enough (git, GitHub CLI, Node.js, npm, plus the profile's own - e.g. .NET SDK, PowerShell 7, Azure CLI, Power Platform CLI, LocalDB, k6) and give the exact install command per OS for anything missing. Runs automatically at the start of ne
dataverse-integrationBuild a sync from a SaaS product to customers' Dynamics 365 / Dataverse environments and Customer Insights - Journeys (multi-tenant app with admin consent and an application user, Web API upserts, contact matching, leads, timeline notes, custom triggers, marketing lists, deep links), tested against
design-systemApply, switch or regenerate a project's design system (the one source of colours, fonts, logos, chart, email, document and video styles for every output), check its contrast in light and dark, and add a new design system to the framework from a Claude Design System artifact. Uses scripts/design-syst
load-testPlan and build a load test per user journey with k6 - scenarios such as a sales rush, the morning-of peak, a check-in burst, live sessions, dashboards and API clients; success criteria (p95, error rate); respecting rate limits; a test data generator with cleanup; a local trial first; and a real-run
mobile-appAdd Android and iOS store apps to an existing web product by wrapping its PWA in a Capacitor shell (one app for every tenant, deep links, offline, push, Wallet, optional Apple Watch companion), with CI builds to Google Play internal testing and TestFlight. Use when the owner asks for native apps, st
new-projectStart a new project with the build framework in the current folder or repository. Asks the owner for the languages (and source language), data residency, stack profile, design system (the look of every output, from design-systems/) and GitHub repository visibility/description, then scaffolds CLAUDE.
ops-readinessPlan and run the Operations readiness wave before the first real users or a big peak - monitoring and alert rules to the owner's email with cost-tiered availability tests (owner chooses the tier), backups (point-in-time retention, long-term copies, storage soft delete) in the chosen region, a restor
owner-checklistWrite or update a checklist of steps only the owner (a person) can do - store submissions, OAuth app registrations, DNS records, payment or legal steps, console settings - with exact click paths, values to enter, where each secret goes (never the secret itself), a status table updated row by row, ke
plan-waveTurn a list of wishes into a wave of 3-6 parallel agent briefs with reserved migration slots, scratch databases, port ranges, file ownership and "small targeted edits" rules, then (on the owner's go) start the agents in the background. Also proposes what to build next in groups (loose ends, operatio
polish-roundRun a polish round after a feature wave - collect issues from screenshots (languages x 390/1440 x light/dark), video recordings (console errors, failed requests, broken images per step), accessibility runs, demo walkthroughs and the owner's "small thing, but..." feedback into one list, assign them w
privacy-reviewPrivacy-by-design (GDPR) and data-sovereignty review of a feature or the whole product - data minimisation, purpose per field, consent, no personal data in logs or analytics, erasure ("delete my data") covering every table and external copy, retention, small-group suppression, signed links, privacy
quality-gateRun the full quality gate and report every step with numbers - tests, release build with the pinned toolchain, front-end builds, vocabulary/translation checker for every configured language, contract tests (changelog, languages, plurals, light/dark theme contrast), the design system check (names, bo
releaseRelease the current work to production with the owner's approvals - commit only relevant files (no AI attribution), push, release from a clean git worktree, pinned-toolchain build, CI green, production migrations as a person's identity (asking before switching the cloud account and restoring it in a
security-reviewSecurity review of new or changed surfaces using the framework's checklist as practised - tenant isolation (IDOR), roles on the server, OAuth state and nonces, emailed-code limits, signed expiring revocable links, webhook signatures and size limits, rate limits shared across replicas, CSP and frame
status-updateSet up (or stop) the recurring status update while background agents run - a session cron or loop every 5 minutes (or the interval in CLAUDE.md) that reports what is running, what finished, what's next and what waits for the owner, one line when nothing changed, and cancels itself when nothing runs.
translateAdd or fix user-facing strings in every configured language (from build/contracts/locales.json) with the CLDR plural forms each language needs, placeholders and markup intact, against the glossary, until scripts/check-vocabulary.mjs passes. Also adds a new language to the project or runs a translati
video-tourPlan and script a tour of short product videos (2-3 minutes per feature area) - part list and order, YAML script per part with narration and Playwright steps against a local run with demo data, subtitles in every configured language, then hand over to the video pipeline in tools/video-tour for recor
whats-newAdd, check or tag What's new (changelog) entries in build/contracts/changelog.json - one plain sentence per user-visible change in every configured language, next free id, valid area, release null until released, then tagged with the release version via scripts/release-tag.mjs. Use when finishing a
write-docsWrite or update project documentation the framework's way - STATE.md (honest status, live version, deployment, data flows, alerts runbook, restore runbook, Unproven table, rules, decisions), ROADMAP.md (waves with status marks, Not doing, Operations readiness, Decisions), runbooks, owner checklists,
a11y-auditorAudits pages against WCAG 2.2 AA with axe-core and probes (light and dark, 1440/390/320), a keyboard walk-through and manual checks, rates findings, and - when its brief allows - fixes them at the shared component with small targeted edits, translated accessible names and contrast-checked tokens, th
feature-builderBuilds one well-scoped feature or fix from an agent brief in a shared working tree, in parallel with other agents - all strings in every configured language, WCAG 2.2 AA, light and dark, privacy by design, data residency, one What's new entry, a STATE.md section, the full quality gate with numbers,
polish-reviewerReviews finished work the way a picky user would - screenshots in several configured languages at 390 and 1440 px in light and dark, video recorder findings, accessibility probes, demo data coherence and owner feedback - lists every issue with evidence, and fixes the small ones it owns (labels, tran
release-checkerRead-only verification before and after a release - checks that the commit holds only finished work, What's new entries exist in every configured language for user-visible changes, no AI attribution or secrets in the diff or commit messages, CI is green for the commit, the gate passed in a clean wor
security-reviewerRead-only security and privacy reviewer. Walks the framework's checklist over the surfaces in its brief - tenant isolation (IDOR), server-side roles, OAuth state, emailed-code limits, signed expiring revocable links, webhook signatures and limits, rate limits shared across replicas, CSP and frame al
translatorTranslates string bundles, What's new entries and other user-facing text into the configured languages it is given, with the CLDR plural forms each language needs, placeholders and markup intact, the glossary's terms and form of address, merging key by key into shared bundles without disturbing othe
node
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
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
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.