mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Hooks, role agents and a code shield that keep Claude Code from quietly breaking what already worked.
Plugin code-guardrails from marketplace alizici in alizici/claude-code-guardrails. Commands read from the repo's manifest — 5 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.
checkCheck that code-guardrails is set up correctly in the current project and explain what it would do. Validates guardrails.json, runs the deletion guard's self-test, lists the code shield's locks, and answers "would the guard allow writing this file / running this command?". Use when the user asks whe
initSet up code-guardrails in the current project. Creates guardrails.json and shield.json for the detected stack (Flutter/Dart, Python, Node, Swift) and installs the code shield's git hooks. Use when the user asks to set up, enable, install or initialize the guardrails, the guard hooks or the code shie
lockLock a piece of critical code with the code shield so it can't change without the user's approval and passing tests. Use when the user asks to lock, protect, freeze or shield a function, rule, file or region (pricing and money rules, protocol encoders/decoders, deterministic math, security checks).
card-auditorAudits a task card BEFORE any code is written. Checks the card's files, decisions, acceptance tests and expected values against the design docs and the codebase, and answers READY or BLOCKED. Read-only.
implementerTurns a card with already-written, red acceptance tests into the smallest correct production code. Only writes production code; the guard denies writes to tests.
reviewerReviews a card's finished work with a clean context. Never sees the implementer's conversation; reads the card, the diff and the tests, measures test strength with mutations, and answers APPROVE or BACK. Read-only in the project.
test-writerWrites a card's acceptance tests BEFORE the implementation exists, so they are red now and turn green when correct code arrives. Only writes test code; the guard denies writes to production code.
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/guard.py" turn
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/deletion_guard.py" turn
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/guard.py" protect
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/deletion_guard.py" before
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/guard.py" record
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/deletion_guard.py" after
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/guard.py" gates
python3 "${CLAUDE_PLUGIN_ROOT}/guardrails/deletion_guard.py" gate
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
CMEM Pro for the Claude app - hooks stream tool use to cmem.ai and inject observations into new sessions and agents
3 skills · 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
AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.
3 skills · 6 commands · 1 MCP server · hooks
Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.
5 commands
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.