mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Pandahrms-specific skills: development workflow orchestration, spec writing, cross-project bridge, system setup, EF Core migrations, React web frontend conventions, code review, and commit workflow
Plugin pandahrms from marketplace pandahrms-skills in pandaworks-software-plt/pandahrms-skills. Commands read from the repo's manifest — 1 month 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.
releaseCut a release of the pandahrms-skills plugin. Triggers on `/release`, "release", "bump version and release", "cut a release", "ship a release", "publish the plugin". Auto-detects the semver bump level from the unreleased changes (new skill / rename / slash-command change = minor; fix/tweak/docs = pa
skill-md-literal-execution-auditManually invoked as `/skill-md-literal-execution-audit <path-to-SKILL.md>` to audit one SKILL.md for literal-execution traps -- hedged language, missing negative constraints, imprecise triggers, implicit sequencing, assumed fallbacks, unclear scope boundaries -- and return rewrites plus a risk summa
branchingALWAYS use when the user needs to create, start, or checkout a new git branch. Prevents dangerous upstream tracking inheritance and enforces folder-based naming conventions. Triggers on any mention of new branch, new feature branch, branching off, or checkout -b
close`/close` -- close a finished piece of Pandahrms work after every card is executed. Verifies every card is already in the per-work `done/` folder and STOPS if any card remains in `active/`; for ticket-driven work invokes `/resolve-ticket` to move the ticket to a resolved, ready-for-release state (sta
code-reviewTriggers on mentions of code review of working-tree changes -- `/code-review`, "review my changes", "check my changes", "review the diff", "review before commit". A diff-scoped LLM-judgment review of the changed files in 3 named modes (standalone | orchestrated | autonomous) -- judgment checklist, f
commitTriggers when the user explicitly requests a git commit of the whole branch''s working-tree changes -- phrases like "commit my changes", "git commit", "ready to commit", "/commit", or "make atomic commits". Branch-scope commit gate -- invokes `/security-review --no-commit` for sensitive changes, aut
discover-projectThe project-queue intake door of the Pandahrms flow. `/discover-project PROJECT_NUMBER [--all]`. Fetches the project's tickets via the workspace-prod MCP `list_tickets` tool -- the pending dev queue by default, every linked ticket with `--all` -- and prints them as a numbered table for the user to p
discover-ticketThe ticket intake door of the Pandahrms flow. `/discover-ticket TICKET_REF` -- a single UUID or human ticket number like T26050092. Fetches the ticket via the workspace-prod MCP `get_ticket` tool, validates and fills gaps in its existing acceptance criteria (never generates it from scratch), and emi
discoverThe free-form intake door for a new feature, enhancement, or bug in Pandahrms. `/discover INTENT`. Turns raw intent into a converged objective plus plain-statement acceptance criteria -- brainstorm mode for a new feature or enhancement, understand-scope-then-root-cause investigate mode for a bug. Wr
ef-migrationsUse when creating, applying, undoing, or troubleshooting Entity Framework Core migrations in any Pandahrms .NET backend project (Performance API, Recruitment API), or when you need to generate SQL scripts from migrations
executeTriggers on requests to run a work card -- `/execute card-NN`, "start card 01", "run the next card", "execute card", "do the next card". Runs that card''s ordered sequence (or the next available card when bare) as a guided, spec-first TDD run with stop-gates, invoking `/lint-gate` and `/code-review`
lint-gateManually invoked as `/lint-gate` (or by an explicit "run the lint gate" / "deterministic guard pass on my changes" mention). A diff-scoped DETERMINISTIC guard runner over the working-tree changes -- project linter on changed files, a Tool Gate (TODO/FIXME/XXX, secrets, leftover-debug, repo-conventio
pr-approver-review`/pr-approver-review PR_NUMBER [fast|deep]` -- senior-approver review of an ALREADY-OPENED GitHub PR by number -- form your own findings and approval gate first (the independent phase always runs in a dispatched subagent), then cross-check the `claude[bot]` review. Reads code at the PR head commit,
pr-review-experimentalReview one GitHub PR with experimental NO-GO / MEL classification, persistent findings and a three-round budget plus one conditional final round. Use /pr-review-experimental with a PR URL, owner/repo#number, or number in a known repository, and optional --fix for requested local repairs. Default is
prTriggers on requests to raise the PR for finished work -- `/pr`, "raise a PR", "raise the pr", "open the PR". Raises the single PR for the whole branch once every card is done -- determines the branch first (asking the user how to branch when the current one is protected), runs `/commit` (clean-tree
resolve-ticketManually invoked as `/resolve-ticket TICKET_NUMBER` (or by an explicit mention of "resolve-ticket" / "resolve this ticket") to move ONE workspace-prod ticket (a UUID or a human number like T26050092) to a resolved, ready-for-release state -- card-less. Fetches it via the workspace-prod MCP `get_tick
security-reviewTriggers when the user requests a security review of code or working-tree changes -- phrasings such as "/security-review", "run security review", "do a security review", "audit this for vulnerabilities", "OWASP pass on this branch", "check this PR for security issues", "auth audit", "authz audit", "
simplifyTriggers when the user explicitly requests a simplification pass on working-tree changes -- phrasings such as "/simplify", "run simplify", "simplify this", "simplify my changes", "do a code-reuse pass", "DRY this up", "make this leaner", "reduce duplication". Dispatches three parallel review subagen
slice`/slice` -- cut agreed work into independently-completable cards. Groups by capability where it helps, with a Collapse Rule for trivial ones. Each card carries the L1 scenarios it covers, its L2 .feature spec file(s), an ordered work-sequence CHECKLIST templated by layer-span AND architecture, a cat
spec`/spec` -- write or update the L1 behaviour spec in a dedicated git worktree of the L1 spec repo (location read from the `Spec repo:` line in the active host''s global instructions; asks the user and saves it there when the location is not stated) from intake output (objective + acceptance criteria
statusThe read-only status step of the Pandahrms flow. Invoked as `/status`, by an explicit request for THIS work's card status -- "status of the cards", "what's left in this work", "what's pending on these cards" -- and by `/execute` when it finishes the LAST card. Scans every card's status, reports the
tool-doctor`/tool-doctor` -- audit the work machine AND the current project for deterministic code-quality guard tools used by the review flow (ripgrep, gitleaks, ast-grep, linters/analyzers, coverage, jscpd). Scans read-only first, prints a readiness table, then OFFERS, with per-item confirmation, to install
verifyManually invoked as `/verify` (or by an explicit mention of "verify" / "run verify"). The single canonical project-scoped deterministic runner for a Pandahrms repo -- full whole-graph build / type-check + full test suite + changed-file coverage existence gate -- emitting one structured PASS/FAIL res
"${CLAUDE_PLUGIN_ROOT}/hooks/run-hook.cmd" session-start.sh
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
Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.
6 skills
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
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
5 skills + 35 modes + Material Passport pipeline. Includes v3.6.7 cross-model audit gate and v3.6.8 generator-evaluator contract.
5 skills · 16 commands · 4 agents · 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.