Claude Code and Codex plugins their authors filed under "workflow". All categories.
Zero-dependency planning + execution with .flow/ task tracking. Three-layer quality (guard + RP plan-review + Codex adversarial). Full-auto, zero questions. Teams auto-parallel, DAG mutation, Codex-driven decisions, auto draft-PR, session summary. 5 prompt templates. 20 subagents, 25+ CLI commands, 23 skills.
25 skills · 18 commands · 24 agents · hooks
Full-stack NestJS + Next.js development workflow — TDD, Figma design integration, browser verification, and doc governance. Combines the full w-task workflow suite (intake → context → plan → impact → UI/Figma → TDD → browser-verify → docs → PR) with the f-* Figma skill suite and Chrome DevTools MCP for localhost UI verification. v0.3.0 workflow v2: per-repo/submodule default branches (main vs master) via workflow.yaml repos[], terminal one-PR-per-repo Phase 6, and w-setup version detection + auto-migration. DEPRECATED as of v0.4.0 — superseded by `nextjs-combo`, which is the maintained successor (official Figma MCP suite, Figma subagents, visual parity gate, infra security review). v0.4.0 only extends the `figma` preset so existing installs keep working with w-task's engine-aware Phase 3; no further features will land here. v0.5.0 is likewise a compatibility bump only — it installs the two skills workflow v3 added (`w-codegraph`, `w-acceptance-verify`) so existing installs keep working with the updated w-task. New work belongs in `nextjs-combo`.
48 skills · 8 commands · 11 agents
Full-stack NestJS + Next.js development workflow. Combines the w-task state machine (intake → context → plan → impact → UI/Figma → TDD → browser-verify → docs → PR) with the official Figma design-to-code suite: figma-context-extractor absorbs the MCP payloads into an exact figma-spec.md, figma-asset-fetcher downloads real icons and images instead of generating them, figma-design-system-mapper resolves Figma values against the project's own tokens and components, and a Phase 3.4 visual parity gate blocks the task until computed styles match the design. Includes Chrome DevTools MCP for localhost verification, the cloud-infrastructure security review checklist, and the terraform MCP server for IaC. Supersedes the deprecated `nestjs-nextjs` preset. v1.1.0 workflow v3: a CodeGraph fast path (`w-codegraph`) lets Phases 0b/2/4 query a code knowledge graph instead of grepping the tree; Phase 4c splits its non-run outcomes and waits out a busy browser MCP before asking the operator rather than self-recording a skip; new Phase 5b (`w-acceptance-verify`) traces every ticket requirement to evidence and compares the UI against the Figma export, fixing findings in at most three committed and re-documented iterations; and Phase 6.3 commits and pushes the final task state into the PR so a finished run never leaves a dirty tree. Requires workflow.yaml v3 — `/w-setup` migrates.
48 skills · 8 commands · 11 agents
An opinionated, disciplined, quality-first software production system. A conveyor that carries value from IDEA to PRODUCTION across role-tuned agents and value-stations (IDEA ▶ ROADMAP ▶ PLAN ▶ EARS ▶ FEATURE ▶ TEST ▶ IMPLEMENT ▶ STORY ▶ SHIP), governed by three pillars — knowledge parity, quality-first with a performance-delta gate, and waste elimination (muda·mura·muri) — under an overarching token-efficiency constraint. Emits markdown; uses the secure and publish companions automatically when present.
19 skills · 13 commands · 32 agents · 1 MCP server · hooks
22 agentes + 30 skills + 13 comandos para uso operacional no Claude Code
30 skills · 13 commands · 22 agents
Full-stack NestJS + Next.js development workflow. Combines the w-task state machine (intake → context → plan → impact → UI/Figma → TDD → browser-verify → docs → PR) with the official Figma design-to-code suite: figma-context-extractor absorbs the MCP payloads into an exact figma-spec.md, figma-asset-fetcher downloads real icons and images instead of generating them, figma-design-system-mapper resolves Figma values against the project's own tokens and components, and a Phase 3.4 visual parity gate blocks the task until computed styles match the design. Includes Chrome DevTools MCP for localhost verification, the cloud-infrastructure security review checklist, and the terraform MCP server for IaC. Supersedes the deprecated `nestjs-nextjs` preset.
46 skills · 8 commands · 11 agents
Unattended implementation: build what was decided, at a cost that tracks what is still unknown. Thirty-five skills, twenty-four principles, nineteen playbooks and three agents, forked from pstack. Reads the rigor signal a spec carries, gates expensive multi-agent work behind one-way doors, and caps every fan-out. Requires ph-lib.
59 skills · 3 agents · hooks
Agents, skills, slash commands, and hooks for Claude Code and Codex: planning, TDD, code and PR review, verification, and UI checks.
32 skills · 13 commands · 14 agents · hooks
Describe what you want in plain language. onestop plans the phases and dispatches the right specialist for each - requirements, design, code, tests, web and app automation, review, ship - stopping at every gate for your approval. A pipeline engine enforces the gates, retry budgets and safety rules in code; technology and version choices and everything that reaches git stay yours.
22 skills · 5 commands · 27 agents · 4 MCP servers · hooks
SGE methodology + shared workflow skills for spec-governed repos — spec-governed implementation, PR merge gates, governance drift sweeps, and an FS AI-governance register.
51 skills · 3 agents · 1 MCP server · hooks
36 interconnected skills (concept → spec → design → build → ship) plus working enforcement: a commit gate that blocks unverified commits, pipeline state at session start, and a pointer to the next skill when one finishes. The skills-only zip ships none of that.
36 skills · 16 commands · hooks
Skills que operan los marcos MFB, MCS, MCC y MCA, más plantillas instalables por proyecto
52 skills
Set the targets. Your agents do the rest. Say F[no] to mostly done.
29 skills · 1 command · 19 agents · hooks
Skills, subagents, and safety hooks for everyday engineering work.
43 skills · 5 agents · hooks
Complete collection of agents, skills, hooks, commands, and rules evolved over 10+ months of intensive daily use
15 skills · 21 commands · 12 agents · hooks
Cross-harness operator layer: 42+ skills, engineering rules, subagents, and memory/learning hooks. Version follows the git commit SHA, so every push auto-updates installs.
43 skills · 6 agents
The Data-First Engineering Protocol as an installable asset set: 28 skills, 14 agents and 14 hooks that put design before implementation and tests before code.
34 skills · 15 agents
Brain-dump → groomed backlog → autonomous execution. 31 skills covering intake, planning, implementation, review, QA, demos, plain-English run briefings, sync meetings, and scheduled reporting.
46 skills · 1 command
Plan → Confirm → Implement → Verify discipline for software work: hooks, skills, review agents and a CLAUDE.md ruleset.
40 skills · 6 agents · hooks
End-to-end GitHub ticket workflow. /kit:architect runs the problem conversation and files lean issues; /kit:design settles the how and writes the plan; /kit:ship-ticket drives TDD through PR with gates you answer, or unattended, parking any ticket whose plan no longer holds. Ships an opt-in PreToolUse hook that holds Rails commits to RuboCop and Brakeman, registered per project.
27 skills · 15 commands · 4 agents
mad (MultiAgent Decanting) — orquestração multi-agente em modo decanting nativo do Claude Code: o Arquiteto coordena especialistas via Agent tool (sem claude -p, sem processos em background), decanting obrigatório ao fim de cada feature, memória institucional em arquivo, trust ladder, guardrails proporcionais ao blast radius, dashboard HTML local cross-platform, OpenTelemetry GenAI e discovery profundo (postura + rigor). Thin wrapper sobre Claude Code; Windows nativo, macOS e Linux.
2 skills · 32 commands · 11 agents · hooks
仕様駆動開発（SDD）の工程エージェント群（要件 → 設計 → 実装 → テスト → レビュー → Git）
25 skills · 20 agents · hooks
Every ClankerKit skill. Prefer a smaller bundle unless you want the whole catalog installed.
44 skills
Trace a failure, separate synthetic green from runtime truth, and leave a blameless handoff.
44 skills
Spend the first hour mapping how a repository actually runs before editing it.
44 skills
Brainstorm, plan and execute with scored tasks, in-session execution, criteria-based review, UI/UX work proven in the running UI and external Antigravity and Codex executors; on Claude Code a cockpit pane shows the session, the run, context against its handoff budget, usage limits and the repo, with Discord notifications and an attribution guard.
25 skills · 19 agents
開発に必要な汎用プラグイン集です。エージェント、スキル、フック、コマンドをまとめて導入し、計画・実装・検証・レビューの流れを揃えます。
21 skills · 10 commands · 9 agents · hooks
Complete collection of agents, skills, hooks, commands, and rules evolved over 10+ months of intensive daily use
11 skills · 15 commands · 13 agents · hooks
The full idea-to-launch skill set for an indie iOS app: ship-app (which skill runs at each phase), new-app (repo template), store-assets (header, search-result and new-device images), first-impression (first-session audit and launch-week plan before 1.0), user-journey (journeys derived from the business, checked in code and walked in the simulator), plus pinned third-party skills for jobs-to-be-done research, acceptance criteria, SwiftUI, TDD, App Store Connect (asc), App Review checks, ASO, launch marketing, 12-month marketing plans, and referral, social, community, creator, AI-search and free-tool marketing.
40 skills
아이디어에서 배포까지 AI 에이전트의 전체 개발 라이프사이클을 구조화. /sdd 자동 파이프라인, /harness 환경 구성, /code-mapper 구조적 컨텍스트 제공(편집 전 실제 호출관계 파악), /pr-converge 외부신호 수렴 루프, /self-improve 자가개선 루프, enforcement 훅(role-gate, tdd-gate, branch-gate, e2e-gate, phase-gate, stop-pipeline 등).
13 skills · 25 agents · hooks