mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Evidence-based workflow with local rules readiness, scoped UI design authority, task recovery and approved flowchart reviews.
Plugin workflow-axa from marketplace gazohhh in Gazohhh/workflow-axa. Commands read from the repo's manifest — today.
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.
code-reviewReview a diff on two separate axes, Standards and Spec. Use for a task's changes, a branch, a pull request or uncommitted work.
codebase-designDeep-module vocabulary for interface and seam decisions. Use when designing or improving a module's interface, choosing where a seam goes or making code easier to test.
codebase-mapDraw an evidence-linked flowchart or codebase map, or the workflow's plan report, with the bundled design guide. Visualization only: never implements or approves work.
domain-modelingSharpen a project's domain terms and decisions. Use when a term, owner or boundary is unclear or changing, or when editing the glossary (CONTEXT.md or GLOSSARY.md) or an ADR.
grillingFind the decisions nobody mentioned, after reading the code. Use through workflow-clarify or to stress-test a plan. Asks only choices that change behavior, scope, safety or compatibility.
implement-specTicket-graph implementation reference only; the workflow never runs it. Use workflow-execute for approved tickets.
plan-reviewReview plans for missing decisions, unnecessary scope, weak verification and AI slop. Never implement or approve for the owner.
resume-taskResume an existing task across sessions. Reconcile records, files, approval and specialist availability before continuing.
retroRun a retrospective over one or many coding sessions.
rule-makerAudit, set up or check a project's agent rules (AGENTS.md, CLAUDE.md) and documentation structure. Use when a readiness check finds gaps, or the user asks to set up, update, audit or check them. Changes a project only after an approved proposal.
simple-englishWrite or rewrite text in plain English that a reader outside the field understands on one read. Use for docs, reports, release notes, error messages and chat replies.
task-statusInspect an existing workflow task and report where it really stands. Reconcile PLAN/TODO/FINDINGS/DECISIONS with the current worktree; do not implement or silently create a new task.
tddTest-first development at agreed seams: red, green, refactor. Use when building a feature or fixing a bug test-first, or when a ticket says TDD.
ui-designUse for any application UI work (behavior-only included), in workflow or ordinary requests, and to build or check a design kit or DESIGN.md. Never write CSS: reuse existing classes and IDs; when nothing matches, ask which existing element to mimic.
workflow-clarifyAgree a task's behavior before planning: trace journeys, callers and failure paths, and grill only for open user decisions. Use for a new goal, a scope change or a contradictory answer.
workflow-deliverCheck a finished or stopped task against its evidence and write the delivery record. Use when implementation ends or someone asks whether a task is really done.
workflow-doctorDiagnose the workflow-axa installation: version, routing, rule coverage, duplicate copies and report assets. Read-only; repairs nothing.
workflow-executeImplement approved tickets, verify and review them, then deliver with evidence. Use only after implementation approval.
workflow-specifyTurn agreed behavior into scoped tickets, dependencies, checks and the plan report for implementation approval. Use after agreement or for a scoped revision.
workflowUse for repository features, fixes, refactors, research, plans, audits and task resumption, including ordinary prompts. Find the task, check rules, route to the phase skill its records require: clarify, specify, execute or deliver. Explain-only stays read-only.
writing-for-agentsWrite and review instructions for agents. Use when creating or editing a skill, AGENTS.md or CLAUDE.md.
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/duplicate_skills.py"
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/phase_gate.py" --hook-protocol=1
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/duplicate_skills.py"
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/truncated_read.py"
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/phase_reads.py"
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/phase_gate.py" --hook-protocol=1
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/phase_reads.py"
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/phase_gate.py" --hook-protocol=1
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/duplicate_skills.py"
python -X utf8 -I "${CLAUDE_PLUGIN_ROOT}/hooks/plain_reply.py"
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
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
Use Alibaba Open Code Review from local Codex.
2 skills
Backend API design, GraphQL architecture, workflow orchestration with Temporal, and test-driven backend development
9 skills · 1 command · 8 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.