mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Specialist agents, grumpy reviewers, and slash commands for project planning, scoping, implementation, and code review. Transforms Claude Code into a multi-agent development team.
Plugin chaos from marketplace chaos in JackScammell/Parliament-Of-Chaos. Commands read from the repo's manifest — 12 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.
/chaos:adr-newScaffold a new Architectural Decision Record (ADR) from a title or ongoing discussion
/chaos:adr-supersedeMark an existing ADR as superseded and link it to its replacement
/chaos:agent-usage-statsShow frequency and effectiveness per agent — input to retirement and optimisation decisions
/chaos:analyse-queriesAnalyse SQL queries and ORM code for performance issues and missing indexes
/chaos:ask-councilAsk the council a question and get a synthesised answer from the appropriate panel of agents
/chaos:changelog-reviewReview new Claude Code changelog entries since last review and propose Parliament features
/chaos:changelogDisplay the Parliament of Chaos version history and changelog
/chaos:ci-watchPoll CI pipelines for the current branch and surface status inside Claude
/chaos:clean-importsRemove unused imports and organise import ordering across changed files
/chaos:commit-and-pushGuided commit + push helper that drafts the commit message and prints the exact git commands for the developer to run — never executes them
/chaos:cost-reportPre-flight and post-flight cost visibility with dry-run, soft caps, and expensive-command warnings
/chaos:coverage-auditRisk-prioritised test coverage analysis — find the most dangerous untested code
/chaos:cut-releaseAutomate version bumping, changelog generation, tagging, and release notes
/chaos:dead-code-sweepFind unreachable code, unused exports, and orphaned files across the project
/chaos:debate-analyticsGenerate analytics dashboard for a debate topic or recent debates
/chaos:debate-replayDeterministically replay a past debate from a session snapshot — regression tests the deliberation engine
/chaos:debate-topicRun a structured multi-agent deliberation on a specified topic
/chaos:decision-reviewRe-evaluate a prior ADR, debate, or council ruling when context has changed
/chaos:docs-auditDetect stale or fabricated documentation after code changes — symmetric to /onboard-codebase
/chaos:env-doctorValidate runtime environment — plugin data dir, hook locations, tool availability, and fallback behaviour
/chaos:explain-agentExplain what a specific agent does and when to use it
/chaos:fast-trackMinimum-review-floor bypass — runs only security + correctness reviewers, never waives them
/chaos:format-codeAuto-detect and run the project's code formatter
/chaos:generate-testsGenerate tests for existing code following project conventions
/chaos:git-workflowAssist with complex git operations — conflicts, cherry-picks, branch cleanup, bisect
/chaos:i18n-auditDrive grumpy-i18n-nitpicker to audit the project for hardcoded strings, missing translations, and locale bugs
/chaos:implement-task-listImplement a roadmap item's tasks with full Parliament oversight
/chaos:incidentStructured incident triage, hotfix coordination, and postmortem generation
/chaos:infra-reviewAudit Dockerfiles, k8s manifests, and docker-compose for best practices
/chaos:lint-fixAuto-detect and run the project's linter(s) with auto-fix
/chaos:list-agentsDisplay all Parliament of Chaos agents grouped by category
/chaos:list-commandsDisplay all Parliament of Chaos slash commands grouped by category
/chaos:mutation-testEvaluate test quality by introducing code mutations and checking if tests catch them
/chaos:onboard-codebaseAnalyse an undocumented codebase and generate comprehensive getting-started documentation
/chaos:parliament-doctorReconcile the command manifest against commands/*.md and the skill registry — report orphans, ghosts, and drift
/chaos:parliament-loopSet up recurring Parliament monitoring using /loop integration
/chaos:parliament-metricsDashboard view — token attribution per agent, session cost, SLO status for background monitors, trend analysis
/chaos:parliament-monitorManage background monitoring agents for continuous code oversight
/chaos:parliament-optimizeAudit agent definitions and recommend effort/model optimisation settings
/chaos:parliament-reviewGrumpy-reviewer review — relevance- and size-tiered by default, PR rounds counted across invocations; --all forces the full 9-member panel (privacy-paranoid joins on personal data, making 10)
/chaos:parliament-webhookConfigure webhook notification endpoints for Parliament events
/chaos:plan-projectGuide a project from idea to structured plan with outline, features, and roadmap
/chaos:plugin-installInstall a community agent plugin from the marketplace
/chaos:plugin-listList all installed agent plugins from the marketplace
/chaos:plugin-upgradeVersion-sync migration helper — bumps plugin.json, marketplace.json, and CHANGELOG together
/chaos:pre-commit-checkAuto-detect and run all CI checks locally before committing
/chaos:project-statusDisplay a dashboard showing the current state of the project
/chaos:readme/chaos:release-notes-draft/chaos:retro/chaos:roadmap-add-item/chaos:roadmap-item-scope/chaos:run-tests/chaos:scaffold/chaos:security-scan/chaos:session-snapshot/chaos:settings-audit/chaos:summon-council/chaos:summon-grumpy-reviewer/chaos:summon-specialist/chaos:telemetry-query/chaos:test-health/chaos:track-debt/chaos:update-dependencies/chaos:update-docs/chaos:versionapi-keeperAPI design specialist. Ensures endpoints are consistent, well-structured and aligned with REST and project conventions. - Task - Agent - SendMessage
backend-goblinBackend performance specialist. Reviews and optimises server-side code, database interactions and infrastructure for speed. - Task - Agent - SendMessage
config-curatorConfiguration management expert. Designs environment configs, secrets handling and feature flag strategies. - Task - Agent - SendMessage
data-warlockDatabase expert. Advises on schema design, query optimisation and indexing strategies. - Task - Agent - SendMessage
deliberation-conductorOrchestrates structured multi-agent deliberation using the Parliament of Chaos deliberation system. Runs debates with parallel execution, convergence detection, and configurable voting systems. Stop: - hooks: - type: command command: '"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_debate_completion.sh' - Tas
dependency-detectiveDependency analysis expert. Investigates upgrade paths, vulnerability chains and license compliance across the dependency tree. - Task - Agent - SendMessage
doc-bardDocumentation specialist. Creates and organises project documentation, turning technical details into clear Markdown. NEVER fabricates information. - Task - Agent - SendMessage
grumpy-accessibility-auditorAccessibility compliance reviewer. Audits for WCAG violations, ARIA usage and inclusive design failures. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-architecture-skepticArchitecture critic. Scrutinises architectural decisions for maintainability, scalability and sustainability. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-budget-hawkCloud cost reviewer. Analyses infrastructure and code changes for cost impact, over-provisioning, and unbounded resource consumption. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-code-reviewerCode quality reviewer. Critiques code for cleanliness, readability and best practices. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-documentation-pedantDocumentation quality reviewer. Scrutinises docs for completeness, accuracy and clarity. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-i18n-nitpickerInternationalisation reviewer. Catches hardcoded strings, missing translations, broken pluralisation, and locale-dependent formatting bugs. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-maintainability-curmudgeonMaintainability watchdog. Evaluates code for long-term maintainability, readability and technical debt. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-performance-trollPerformance critic. Analyses code for performance and resource efficiency, identifies bottlenecks and recommends optimisations. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-privacy-paranoidPrivacy and data protection reviewer. Audits code for PII exposure, GDPR/CCPA compliance, consent handling, and data retention violations. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-security-nagSecurity nagger. Scrutinises code for security risks and insists on proper defences. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-standards-enforcerStandards compliance checker. Verifies code conforms to project conventions and documented standards. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
grumpy-testing-tyrantTest coverage reviewer. Enforces comprehensive testing and rejects inadequate test suites. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
migration-monkDatabase and code migration specialist. Plans safe schema changes, data transformations and rollback strategies. - Task - Agent - SendMessage
observability-oracleObservability specialist. Designs logging standards, metrics collection, distributed tracing and alerting strategies. - Task - Agent - SendMessage
package-wizardDependency health specialist. Audits and optimises project dependencies for security, updates and minimal bloat. - Task - Agent - SendMessage
pipeline-engineerCI/CD and deployment advisor. Reviews and optimises build, test and deployment pipelines. - Task - Agent - SendMessage
project-oracleProject planning interviewer. Guides users through structured Q&A to create project definitions with outlines, features and roadmaps. Welcome! I'll guide you through structured project planning. Let's start — what is the core idea or problem you want to solve? - Task - Agent - SendMessage
refactor-rangerCode refactoring specialist. Identifies code smells, applies refactoring patterns and plans incremental improvements. - Task - Agent - SendMessage
resilience-tamerFault tolerance specialist. Assesses and improves system resilience for external dependencies, distributed systems and failure scenarios. - Task - Agent - SendMessage
scope-weaverRoadmap item scoping specialist. Breaks down roadmap items into detailed specifications and actionable task lists. I'll help scope this roadmap item into a detailed specification. Which roadmap item would you like to break down? - Task - Agent - SendMessage
security-knightSecurity guardian. Assesses authentication, authorization, data protection and overall security posture. - Task - Agent - SendMessage
senior-councilCoordinator meta-agent for multi-disciplinary work. Orchestrates specialist agents and grumpy reviewers for complex tasks spanning multiple domains. - Task(backend-goblin) - Task(ui-ux-guru) - Task(data-warlock) - Task(security-knight) - Task(system-architect) - Task(test-prophet) - Task(pipeline-en
system-architectArchitecture advisor. Designs and evaluates system architecture, domain boundaries, scalability and maintainability. - Edit - Write - NotebookEdit - Bash - Task - Agent - SendMessage
task-executorTask tracking utility. Handles safety checks, task loading, progress tracking and documentation. Used by senior-council during task implementation. - Read - Write - Edit - TaskCreate - TaskUpdate - TaskList - TaskGet - Task - Agent - SendMessage
test-prophetTesting strategist. Plans and reviews test suites, identifies coverage gaps, and provides example tests. - Task - Agent - SendMessage
ui-ux-guruUI/UX design advisor. Reviews interfaces for brand compliance, accessibility and usability. - Task - Agent - SendMessage
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/notify_teams.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/notify_teams.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/notify_teams.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/notify_teams.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/notify_teams.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/log_event.sh
"${CLAUDE_PLUGIN_ROOT}"/src/hooks/notify_teams.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
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
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.