workflow-kit

Plugin workflow-kit from marketplace skill-forge in lucasAguiar11/agent-skills. Commands read from the repo's manifest — 12 days ago.

Skills (17)

• clean-comments

  Limpa comentários ruído do código: remove `// GET /path`, banners de seção, JSDoc que só repete o nome, paráfrase da linha seguinte, TODO sem dono e código morto comentado. Mantém só gotcha/invariante e header curto de módulo (papel + why). Default = remover. Roda no fim da entrega de feature (Post-

• code-review-and-quality

  Code review multi-eixo (correção, legibilidade, arquitetura, segurança, performance) de qualquer mudança antes do merge — Pull/Merge Request (GitHub ou GitLab) ou diff local sem PR aberto. Use sempre que o usuário pedir para revisar um PR/MR, analisar mudanças de código, fazer code review, avaliar u

• commit

  Cria commits seguindo o padrão Conventional Commits em português. Divide mudanças em commits pequenos e atômicos. Use quando quiser fazer commit das alterações.

• create-architectural-decision-record

  Create an Architectural Decision Record (ADR) document for AI-optimized decision documentation.

• create-implementation-plan

  Create a new implementation plan file for new features, refactoring existing code or upgrading packages, design, architecture or infrastructure.

• diff-review

  Revisão de código em pt-BR do diff desde um ponto fixo (commit, branch, tag ou merge-base) em dois eixos: Padrões (a mudança mantém a estrutura atual e segue as convenções que o repo já usa?) e Enxugamento (ponytail-review: o que dá pra cortar?). Só reporta o que tem evidência no código; zero findin

• fast-subagent-protocol

  Decide quando delegar e coordena subagents com contexto pronto, checkpoints por papel, cancelamento seguro e validação centralizada. Use antes de lançar Workers, Scouts, Reviewers ou Validators.

• feature-delivery

  Orchestrate lightweight feature delivery for new or existing projects using feature briefs, optional PRDs, optional ADRs, implementation plans, review gates, ownership boundaries, wave-based subagent launch/handoff, and Integration Coordinator execution for parallel human/AI work.

• prd

  Generate high-quality Product Requirements Documents (PRDs) for software systems and AI-powered features. Includes executive summaries, user stories, technical specifications, and risk analysis.

• quick-plan

  Investigate an ambiguous bug/feature request, converge on a design with the user, then hand off a plan for approval — no docs/features.md, ADR, or plan file required. Use when the request has more than one plausible design (e.g. "we have two ways X can happen, what should we do") and a wrong guess w

• review-plan

  Revisa documentos de plano/implementação em loop até não encontrar problemas. Verifica consistência, ordem, referências cruzadas, e completude. Use quando quiser validar um plano antes de executar.

• safe-handoff-execution

  Executa handoffs cross-repo com contrato verificável, escopo rígido e proteção contra sobrescrita acidental. Use ao consumir um handoff de outra aplicação.

• simplify

  Review changed code for reuse, quality, and efficiency, then fix any issues found. Use after implementing a feature, before commit/PR.

• supersede-feature

  Fold one or more old/superseded features into the new feature that replaced them — condense the relevant history into the successor's brief, mark the old ones deprecated with superseded_by, and git rm their dead plans (which stay recoverable in git history). Use when features are obsolete, replaced,

• test-guide

  Audit and design useful tests for this project. Use when writing tests, reviewing test quality, deciding unit vs integration vs E2E, avoiding placebo tests, or when feature-delivery review/execute touches domain rules, persistence, validation, or API behavior.

• update-implementation-plan

  Update an existing implementation plan file with new or update requirements to provide new features, refactoring existing code or upgrading packages, design, architecture or infrastructure.

• verification-before-completion

  Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Subagents (8)

• adr-correlator

  Given a scope/topic, scans all ADRs reading ONLY their frontmatter (scope, tags, status, title) and returns a ranked shortlist of prior decisions that likely bear on the current work — without loading any ADR body. Use in plan mode (Default Flow step 4) so the planner links related decisions instead

• adr-reader

  Reads one ADR by id or path and returns a compact digest — status, the decision, and its consequences — without loading the full ADR. Use when a plan or review needs to honor a prior decision but not re-read its whole context.

• diff-reviewer

  Fast single-pass review (haiku) of the diff since a fixed point, on the Patterns and Trimming axes of the diff-review skill. Read-only. Receives the point (commit, branch, tag) and returns the skill's `Saída` section verbatim in pt-BR.

• feature-index-reader

  Reads docs/features.md (the feature index) and returns a compact view of related prior features — their status, owner, and any dependency/convention the current feature should inherit — without loading each feature's full brief. Use in plan/triage to spot dependencies and reuse before drafting.

• feature-reader

  Reads a feature brief or PRD (docs/features/<FEATURE-ID>.md) and returns a compact digest — problem, scope, requirements, acceptance criteria, open decisions — so the caller can plan or review without loading the full document.

• plan-detail-reader

  Reads ONE Task block from docs/plans/<FEATURE-ID>-plan.md and returns its full detail (objective, files, steps, verification, stop conditions) ready to embed in a Worker subagent prompt. Use during execute to hand a single task to a Worker without loading the whole plan.

• plan-reader

  Reads docs/plans/<FEATURE-ID>-plan.md and returns a compact digest — Goal, Task list, Decision State, Validation status, and Traceability gaps — without loading the whole plan into the caller's context. Use before review/execute, or whenever a stage needs the plan's shape but not its full body.

• task-validator

  Adversarial validator for ONE completed workstream during execute. Receives the Task block and the Worker's handoff, and tries to REFUTE completion — unmet steps, failing verification, out-of-scope writes, weakened tests. Returns `validated` or `refuted` with evidence. Independent of the Worker and

Hooks (1)

• SessionStart

  bash "${CLAUDE_PLUGIN_ROOT}/hooks-handlers/session-start.sh"

Similar plugins