mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Engineering workflow skills and review agents: refine a plan to convergence through a reviewer panel, staff its execution with right-sized subagent clusters, weigh design options, port reference algorithms rigorously, audit duplication, and write up a session.
Plugin xander-local from marketplace xander-plugins in Xander-git/xander-plugins. Commands read from the repo's manifest — 1 month ago.
dedupe-scannerAudit a codebase for extractable duplication and propose targeted refactors. Use this whenever the user asks for a code review focused on duplication, DRY violations, repetition, magic values, code smells, cleanup, or extraction opportunities. Also use proactively when the user shares multiple modul
execute-plan-orchestrationUse when planning subagent execution of a multi-task implementation plan — deciding how many subagents to dispatch, how to group tasks, which model each gets, and where review gates go. Turns a task list + dependency DAG into clusters sized for one agent each. Trigger after writing-plans, before dis
plan-refineryUse when a design spec and implementation plan are drafted and should be iterated to convergence before execution — dispatches a reviewer panel (general, data-flow, simplicity, plus specialists) in a loop, merges severity-tagged concerns into a provenance-locked ledger, applies plan fixes, gates all
porting-a-reference-algorithmA checkable procedure for transcribing an external algorithm (a paper, or a reference implementation) into a codebase. Use whenever matching code to an outside source — before making any claim about what that source does. Covers assembling references, citing file:line, diffing not inspecting, pinnin
present-solutionsPresent multiple candidate solutions to a problem or open question, each with background, tradeoffs (accuracy, implementation cost, complexity, technical debt, plus domain-specific axes), failure modes, and a reasoned recommendation. Use this skill whenever the user is weighing design choices, compa
summarize-sessionGenerate a shareable, evidence-gated write-up of the CURRENT Claude Code session — what was asked, how it was steered, what was built, and what was learned. INVOKE ONLY ON EXPLICIT REQUEST — never proactively or automatically. Trigger only when the user explicitly asks for it, e.g. "summarize this s
data-flow-reviewerFixed panel member of the plan-refinery loop. Traces every user-facing flow in a design spec end-to-end — entry point, validation, state, computation, persistence/output, result surfaced back — and hunts broken chains, boundary contract mismatches, orphaned inputs/outputs, and stale-state hazards. D
general-reviewerFixed panel member of the plan-refinery loop. Reviews a design spec + implementation plan for spec↔plan traceability (coverage gaps, scope creep, contradictions), feasibility against the real codebase, and failure modes (edge cases, partial failures, idempotency). Dispatch via the plan-refinery skil
implementation-test-reviewerUse when a logical chunk of functionality is complete and you need a review that validates implementation correctness AND whether its tests genuinely prove that correctness. Hunts bugs and edge cases, then checks the test suite for false greens — tests that pass without exercising the code paths the
plan-reviewerUse when an implementation plan has been drafted and needs critical review for feasibility, correctness, and optimality before implementation begins. Validates logic against the real codebase, verifies external API claims, hunts race conditions, and challenges unnecessary complexity. Spawn after a p
simplicity-reviewerFixed panel member of the plan-refinery loop — the counterweight that prunes instead of adds. Reads ONLY the spec, plan, context brief, and ledger (no codebase traversal) and asks what can be removed while still satisfying the spec. May challenge the spec itself against its Objective & Non-goals; al
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
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
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
Backend API design, GraphQL architecture, workflow orchestration with Temporal, and test-driven backend development
9 skills · 1 command · 8 agents
Use Codex from Claude Code to review code or delegate tasks.
3 skills · 8 commands · 1 agent · 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.