Workflow plugins

Autonomous task-driven loop that works through TASKLIST.md independently

1 command · hooks

클로드 코드 계획을 태스크로 변환하여 멀티 서브에이전트로 실행하는 워크플로우 플러그인

1 skill · 1 command

Git workflows: reviewed commits that split unrelated changes and carry the conversation's reasoning into the message, and a pull request flow with tests, scoped lint and type checks, a review gate, CI-gated merge and resync. Never bypasses hooks or force-pushes.

2 skills

Run several Claude Code sessions on one git repository without them colliding — session-start probe hook, isolated-worktree and shared-checkout modes, claim files with liveness, the FREEZE?/THAW and resource-lock vocabulary, addressing failures, and recovery commands.

1 skill · hooks

Composable pipeline (intake → repo-detect → repo-fetch → worktree → agent-scan → task-plan → dispatch) connected by a single state.yaml contract, with a per-session harness-events.log feeding the harness-engineering improvement loop. Workflow-agnostic: no role taxonomy, no ordering invariants, no Slack/tmux coupling.

1 skill · 1 agent

Yeet tasks from Claude Code to the Grok Build CLI through a standardized headless-run API: single runs, parallel agent fan-out, session resume, and structured JSON result envelopes.

1 skill · 1 command

Smart conventional commits with automatic change analysis and splitting

1 skill · 1 agent

Adversarial second-opinion review of a finished plan, design doc, or proposal, run before it reaches the human reviewer. Ships as a subagent plus a /devils-advocate command to invoke it explicitly.

1 skill · 1 agent

Rename plan files from random names to meaningful titles extracted from content

1 skill · 1 command

Pull and merge request workflows for GitHub and GitLab

2 commands

立法の支援。GitHub issue での議論を、実装根拠になる spec-kit の仕様(specs/<機能>/)まで持ち上げる。grill-issue が曖昧な issue を質問攻めで詰め、spec-draft が spec-kit のコマンドで仕様を起草して仕様の PR にする。承認は必ず人間。自律ループを回さなくても単体で使える。

2 skills

人間が承認した spec-kit の仕様を起点に拾い上げ→実装→司法検証→PR を回す自律ループ(HOTL)。loop-ready の issue を PR まで進め、必ず PR で停止する。dev-crew と review-judge が前提。

1 skill · 1 command

Operational safety rails for long-running operations. Pre-flight algorithm fitness CLI (working set vs RAM ceiling, with concrete alternatives), in-flight heartbeat helper + cron watcher (stale + ETA overrun detection), Discord + gh issue auto-emit. Human-in-loop only — no auto-kill. Built after a 23h sunk-cost incident where philosophy-level early-bleeding-detection lived only in CLAUDE.md memory, not in structural rails.

1 skill · hooks

Drive the quest CLI: file, plan, record and complete tracker work in a Quest workspace. Federated: pinned to the quest-cli release tag so the skill cannot drift from the CLI it describes.

2 skills

Quality gate for Claude Code skill authoring: whenever a skill is created or edited, the work runs at maximum reasoning effort — via a bundled max-effort agent inheriting the session's model — and is cross-checked against the skill-creator and skill-development skills when installed. Auto-engages on any SKILL.md work.

1 skill · 1 agent

Drive one small task from idea to merged PR through ten fixed steps with approval gates: statement, stories, mockup, plan, TDD, review panel, cross-agent review, live pass, PR, summary.

1 skill · hooks

Git workflow: Conventional Commits formatting that resists the common CI-guard failure modes, and PR creation that detects the remote host (GitHub / Azure DevOps Repos / GitLab) instead of hardcoding one. Pairs with a project-local writing-style fork for tone.

2 skills

End-to-end value-delivery lifecycle: a /deliver command + skill that drives one body of work through plan -> adversarial plan review -> subagent execution -> completion gate -> adversarial code review -> land -> retrospective, composing superpowers + docs + retrospective skills. Project-specific plan/doc/checklist steps bind per-repo via .claude/delivery.local.md; no project-internal names are committed.

1 skill · 1 command

The PR lifecycle, powered by the gh CLI. /bymax-pr:push ships work safely (feature branch — never the default branch — staged set respected, complete Conventional-Commits message, push with upstream; `pr` token opens a fully-described PR). /bymax-pr:babysit-pr shepherds the open PR to merge-readiness: resolves conflicts, polls CI at a cadence matched to the CI's real duration (270 s floor), classifies real vs flaky failures, fixes real ones locally, triages bot review comments, and notifies you when green. Never merges, never pushes to the base branch. Depends on gh + git; never bundles them.

1 skill · 1 command

Instinct-based learning system — observes sessions, creates confidence-scored atomic behaviours, and evolves them into skills

1 skill · 1 agent

1 command + 1 agent (/promote-memory). Scans project memory/errors.md and memory/learnings.md, promotes qualifying status:new entries to permanent memory (MEMORY.md or ~/.claude/CLAUDE.md). Supports dry-run and custom install target.

1 command · 1 agent

Jira addon: push a ticket draft to Jira as an issue. Ships the Atlassian MCP server.

1 skill · 1 MCP server

Run a multi-session job as its orchestrator: plan, brief worker sessions, monitor, verify their claims, and loop the human in only for decisions.

2 skills

Keep plans and delivered work aligned with user-approved anchors.

1 skill · hooks

reviewing-ceremony: per-category Jev review of sprint plans, CI workflows and agent instructions; locates items that match a ceremony pattern.

1 skill · 1 agent

Tool- and project-agnostic ticket workflow — discovery, drafts, backlog grooming, tracked development, and handoffs. Adapts to available integrations or local files. Work-source writes require explicit confirmation.

2 skills

PR review skills: comprehensive scoped review (review-pr) and critical processing of incoming review feedback (handle-code-review).

2 skills

Model Python domain states and keep domain decisions predictable at side-effect boundaries.

2 skills

A thin coordinator session for parallel Claude Code sessions: a daily ledger, scoped briefs, a findings protocol, and one verified merge at a time.

1 skill · hooks

Agent skills from Kale: purge audits, prunes, writes and reviews agent skills and agent context; team-lead plans from your tracker and runs worker subagents, one PR each.

2 skills