mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Bootstrap new repos or improve existing ones for agentic development: map, guardrails, feedback loops.
Plugin agentic-harness from marketplace agentic-development in jaycorpstudios/agentic-development. 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.
agents-md-authorAuthor or repair a repository's AGENTS.md (and its CLAUDE.md mirror) using the canonical sections — Mission, Structure, How to work here, Start here, Skills, Non-negotiables, Preferred workflow, If information is missing. Use when creating, reviewing, or improving a repo's agent entry point.
define-new-projectInterview the user and scaffold a NEW repository for full agentic development — AGENTS.md/CLAUDE.md, a docs/ skeleton, .claude/ hooks + settings, .nvmrc, and the MUST patterns (typed errors + wide events, plus a typed SDK when there's a TS API). Use when starting a new project or when the user runs
design-system-storybookDecide whether a design system is warranted and, if so, structure it as a shared UI package with design tokens and Storybook for component documentation, plus a "where does this component go?" decision rule. Use when a frontend reuses UI across views or apps.
documentation-styleApply the repo documentation conventions — short single-topic files, README-table indexes, links reachable in ≤2 hops, and Mermaid-first for flows and architecture. Use when writing or restructuring docs, adding a doc folder, or diagramming a flow.
error-handling-neverthrowImplement typed error handling with neverthrow Result/ResultAsync and Tagged Errors (a _tag discriminated union), converting throws into values and resolving at the boundary with .match. MUST for new projects; opt-in (tracked tech-debt) for existing ones. Use when adding or standardizing error handl
improve-existing-projectAudit an EXISTING repository against the agentic-readiness criteria and produce a prioritized remediation plan, then apply guardrails and map first. Typed errors and wide events are offered as opt-in (tracked as standardization tech-debt) — never a wholesale rewrite. Use when improving an existing r
monorepo-turboDecide whether a monorepo is warranted and, if so, set it up with Turbo + pnpm — apps/ and packages/ layout, workspace config, task pipeline, and dependency rules. Monorepo is suggested when warranted, never mandatory. Use when structuring multiple deployables that share code.
repo-structureApply domain-based file/folder structure and naming conventions for frontend (Next.js / TanStack Start / SPA) and backend (NestJS / Express), or propose an incremental improvement for an existing repo with poor structure. Use when laying out a new project or restructuring an existing one.
typescript-sdk-generationSet up a typed TypeScript SDK as the single contract between an API and its TS consumers — generate an OpenAPI doc, run openapi-typescript to emit types, then hand-write a Result-wrapped client and zod schemas. Wire the Turbo task graph so types regenerate after the doc. Use when a TS API needs a ty
wide-events-loggingImplement Wide Events logging — one canonical, enriched event per request/job using AsyncLocalStorage, logged once at the end with outcome, duration, and tagged error. MUST for new projects; opt-in (backfill as tech-debt) for existing ones. Use when adding or standardizing structured logging in a se
/agentic-harness:agentic-doctorScore the current repo's agentic-readiness against the criteria and list the top gaps.
/agentic-harness:agentic-improveAudit an EXISTING repository and produce a prioritized plan to make it agentic-ready.
/agentic-harness:agentic-initInterview and scaffold a NEW repository so it is ready for full agentic development.
harness-auditorRead-only auditor that scores a repository's agentic-readiness against the criteria and emits prioritized findings. Use during /agentic-improve or /agentic-doctor.
"${CLAUDE_PLUGIN_ROOT}"/templates/claude/hooks/block-env-files.sh
"${CLAUDE_PLUGIN_ROOT}"/templates/claude/hooks/block-node-modules.sh
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
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
Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.
5 commands
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.