mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Self-improving agent rig: codifies product and engineering discipline into composable workflows from brainstorm to ship. Compounds knowledge, generates domain agents, monitors its own docs, and surfaces conservative update drift. Orchestrates Claude, Codex, and Oracle through 6 agents, 66 commands, 28 skills, 0 MCP servers. Factory substrate: CXDB turn DAG, scenario bank with satisfaction scoring, evidence pipeline, agent capability policies. Companions: interspect, interphase, interline, interflux, interpath, interwatch, interslack, interform, intercraft, interdev, interpeer, intertest.
Not in a marketplace yet, so it loads for one session with --plugin-dir.
Plugin clavain in mistakeknot/Clavain. Commands read from the repo's manifest — 3 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.
bug-reproduction-validatorReproduce and validate a bug report to confirm whether the reported behavior is an actual bug.
code-review-disciplinecodex-delegatedata-migration-expertdispatching-parallel-agentsengineering-docsexecuting-plansfile-todosfull-stopgalianagardenerinterop-with-gsdinterop-with-superpowersinterserve-enginelanding-a-changelaneplan-reviewerpr-comment-resolverproject-onboardrefactor-safelyremontoireshipsubagent-driven-developmentui-polishupstream-sync-engineusing-clavainusing-tmux-for-interactive-commandswriting-plans/clavain:bead-sweepScan open beads for stale trackers already implemented — deterministic checks then LLM verification.
/clavain:blast-radiusFind what a change could break somewhere else before it ships, and prove the one fact it is safe because of by running real code
/clavain:brainstormCollaborative brainstorm — assess clarity, research repo, explore approaches, capture design. Auto-handoff to /write-plan.
/clavain:campaignOrchestrate epic execution — read children, topo-sort by dependencies, dispatch phases sequentially with parallel features
/clavain:changelogCreate engaging changelogs for recent merges to main branch
/clavain:clavain-doctorQuick health check — verifies MCP servers, external tools, beads, and plugin configuration without making changes
/clavain:clavain-helpShow Clavain commands organized by daily drivers first, then by workflow stage
/clavain:clavain-initScaffold .clavain/ agent memory directory in the current project
/clavain:clavain-reviewPerform exhaustive code reviews using multi-agent analysis and deep inspection
/clavain:clavain-statusUnified status view across Clavain, artifact generation, doc drift, and coordination
/clavain:clodex-toggleToggle interserve execution mode
/clavain:codex-bootstrapBootstrap or check Clavain Codex installation health (install + wrapper sync + doctor)
/clavain:compoundDocument a recently solved problem to compound your team's knowledge
/clavain:correctFind the mistakes agents keep repeating in this repo and make each one impossible, at the highest level that works
/clavain:create-agent-skillCreate or edit Claude Code skills with expert guidance on structure and best practices
/clavain:debateRun a structured Claude↔Codex debate before implementing a complex task
/clavain:deepenRewrite an ask as a deeper version, not a bigger one: name the top practitioner's method, the rawest source, the real instruments and the external evidence that would prove the result
/clavain:describe-prGenerate PR title and description from current branch commits
/clavain:distillSynthesize accumulated docs into categorized solutions and generate missing SKILL-compact.md files
/clavain:execute-planSlash-command entry that enforces the pre-execution gate, records the phase transition, then runs a written plan via the executing-plans skill. Use the executing-plans skill directly when no gate or phase tracking is wanted.
/clavain:fixbuildRun build, capture error, fix it, re-run — fast loop for type errors and build failures without heavyweight debugging
/clavain:freezeRestrict file edits to declared paths — scope-lock for parallel-session safety
/clavain:galianaDiscipline analytics — view KPIs, report defects, or reset cache
/clavain:generate-commandCreate a new custom slash command following conventions and best practices
/clavain:goal-formCollaborative goal-formation ritual: research-first inter-elicitation → charter → stakes-routed review → mint (ic goal) → /goal handoff
/clavain:grill-meInterview the user relentlessly about a plan or design until every branch of the decision tree is settled
/clavain:handoffGenerate a concise session handoff summary as a dated markdown file, grounded in intercore run state
/clavain:heal-skillFix incorrect SKILL.md files when a skill has wrong instructions or outdated API references
/clavain:interserveRun Codex-first execution flow for larger scope work
/clavain:landRun the landing workflow for trunk-based handoff
/clavain:migration-safetySlash-command entry that fans out three parallel agents for a full pre-deploy safety assessment of a DB migration or data change. For a single-reviewer validation of migration code against production data, use the data-migration-expert skill instead.
/clavain:model-routingToggle subagent model routing between economy (smart defaults) and quality (all opus) mode
/clavain:next-goalGenerate a Next-goal block (2-4 leverage-ranked candidates + recommendation + ready-to-paste /goal text) — invoke manually after a /goal completes or whenever you want a fresh set of candidates
/clavain:peersRead-only viewer for detected peer agent rigs (superpowers, GSD, compound-engineering). Lists detection state and recommended bridge skills. Never makes changes.
/clavain:plan-reviewHave multiple specialized agents review a plan in parallel
/clavain:pr-triageTriage all open PRs — batch by theme, review with parallel agents, generate report, walk through decisions. With a PR reference, run one follow-up pass on your own PR after a watcher update.
/clavain:prototypeBuild disposable code to answer one design question, record the evidence and answer, then remove or quarantine the experiment
/clavain:quality-gatesGate orchestrator — prepares diff, delegates review to flux-drive, enforces pass/fail gate
/clavain:recallSearch across all knowledge systems — curated docs, memory files, legacy entries
/clavain:refactorPlan and execute safe refactors with explicit risk controls
/clavain:reflectCapture sprint learnings and register the reflection artifact
/clavain:remontoireOperate the Remontoire portfolio agency across hosts with explicit approval and receipt boundaries
/clavain:repro-first-debuggingDisciplined bug investigation — reproduce first, then diagnose. Composes debugging agents.
/clavain:resolveResolve findings from any source — auto-detects TODOs, PR comments, or todo files, then resolves in parallel
/clavain:retroRead a session, a PR with its session, or a week of merged PRs and propose ranked changes to the environment that would have stopped repeated mistakes
/clavain:review-disciplineRun disciplined code review and feedback triage
/clavain:review-docQuick single-pass document refinement — assess clarity, score quality, fix issues, offer iteration
/clavain:routeUniversal entry point — discovers work, resumes sprints, classifies tasks, and dispatches to /sprint or /work
/clavain:setupBootstrap Clavain for the active runtime (Codex or Claude Code) and verify health
/clavain:show-me-your-workKeep a reviewable decision trail for long-running or unattended work, one TSV row per decision
/clavain:smoke-testRun smoke tests against a running app — detect dev server, walk critical user journeys, report results
/clavain:sprint-dagVisualize sprint execution as a DAG — shows phases, dispatches, and artifacts
/clavain:sprint-statusDeep sprint workflow scan — session continuity, pipeline, plan, beads, recommendations.
/clavain:sprintPhase sequencer — brainstorm, strategize, plan, execute, review, ship. Use /route for smart dispatch.
/clavain:strategyStructure brainstorm output into a PRD with features, create beads for tracking, and validate before detailed planning
/clavain:tddRun disciplined test-driven development (RED-GREEN-REFACTOR) for a task
/clavain:todosCreate and maintain file-based todo tracking
/clavain:triageTriage and categorize findings for the CLI todo system
/clavain:unfreezeLift the /freeze scope lock — restore full edit permissions
/clavain:update-checkConservative update check — reports available Sylveste/Clavain/Codex updates without changing anything
/clavain:upstream-syncCheck upstream repos (beads, oracle, superpowers, compound-engineering) for updates and generate upgrade checklist
/clavain:verifyRun completion verification before claiming work is done
/clavain:wayfindChart or work a decision map in beads for an effort whose destination is known but whose route spans many sessions
/clavain:waymapChart or work a project's Now / Next / Later roadmap in beads, as bets the coordinator commits, defers or drops
/clavain:workExecute work plans efficiently while maintaining quality and finishing features
/clavain:write-planCreate detailed implementation plan with bite-sized tasks
architecturecodex-integrationcommands-referencedevelopmenthooks-referenceinterface-contractsskills-reference${CLAUDE_PLUGIN_ROOT}/hooks/context-gateway.sh
${CLAUDE_PLUGIN_ROOT}/hooks/session-start.sh
${CLAUDE_PLUGIN_ROOT}/hooks/peer-telemetry.sh
${CLAUDE_PLUGIN_ROOT}/hooks/release-canary-check.sh
${CLAUDE_PLUGIN_ROOT}/scripts/remontoire-attention.sh
${CLAUDE_PLUGIN_ROOT}/hooks/context-reset-post.sh
${CLAUDE_PLUGIN_ROOT}/hooks/guard-plugin-cache.sh
${CLAUDE_PLUGIN_ROOT}/hooks/security-triage-hook.sh
${CLAUDE_PLUGIN_ROOT}/hooks/gauge-gate-executor-spawn.sh
${CLAUDE_PLUGIN_ROOT}/hooks/context-reset-pre.sh
${CLAUDE_PLUGIN_ROOT}/hooks/peer-routing-telemetry.sh
${CLAUDE_PLUGIN_ROOT}/hooks/auto-publish.sh
${CLAUDE_PLUGIN_ROOT}/hooks/bead-agent-bind.sh
${CLAUDE_PLUGIN_ROOT}/hooks/agents-md-refresh.sh
${CLAUDE_PLUGIN_ROOT}/hooks/catalog-reminder.sh
${CLAUDE_PLUGIN_ROOT}/hooks/validate-plugin-edit.sh
${CLAUDE_PLUGIN_ROOT}/hooks/temple-invariant.sh
${CLAUDE_PLUGIN_ROOT}/hooks/context-reset-post.sh
${CLAUDE_PLUGIN_ROOT}/hooks/auto-stop-actions.sh
${CLAUDE_PLUGIN_ROOT}/hooks/dotfiles-sync.sh
${CLAUDE_PLUGIN_ROOT}/hooks/auto-push.sh
${CLAUDE_PLUGIN_ROOT}/hooks/gate-calibration-session-end.sh
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Harness-native ECC operator layer - 71 agents, 314 skills, 98 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
200 skills · 98 commands · 71 agents · 1 MCP server · hooks
Automated code review for pull requests using multiple specialized agents with confidence-based scoring to filter false positives
1 command
Tenant-isolated AI teams that plan work, coordinate tasks, recall approved context, and return evidence-backed results
6 skills · 6 commands · 4 agents · 1 MCP server
Use Alibaba Open Code Review from local Codex.
2 skills
Investigate incidents, review code and execute UAT for release readiness, scan code for vulnerabilities, and run penetration tests with AWS DevOps Agent and AWS Security Agent.
13 skills · 9 commands · 1 MCP server
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.