PWDEV Code — Spec-Driven Development

Plugin pwdev-code from marketplace pwdev-claude-marketplace in pwdev-solucoes/pwdev-claude-marketplace. Commands read from the repo's manifest — 1 month ago.

Skills (2)

• skill-frontend-design

  Design precise, production-grade enterprise UI — dashboards, admin panels, SaaS interfaces, and data-heavy applications. Use this skill for professional software interfaces that need Jony Ive-level craft: clean, minimal, functional, with intentional personality. Do NOT use for creative landing pages

• skill-user-stories

  Write and refine high-quality user stories — INVEST criteria, "As a X, I want Y, so that Z" format, Gherkin acceptance criteria when applicable, definition of ready, anti-patterns, and a review checklist. Use when creating or reviewing user stories, PRD section 6, backlog items, or acceptance criter

Slash commands (23)

• /pwdev-code:audit

  Query the audit trail database for events, decisions, artifacts, and statistics. Export full PDF report.

• /pwdev-code:codex

  Delegate an implementation task to the Codex CLI, then critically review the resulting diff

• /pwdev-code:delegate

  Pick the best external coding CLI for the task (codex, opencode, kimi, gemini, or kiro) and delegate with the standard safety protocol

• /pwdev-code:design

  Run the design phase to produce spec.md with architecture decisions

• /pwdev-code:discover

  Run the discovery phase to gather requirements through interviews and research

• /pwdev-code:execute

  Execute planned tasks by delegating to the executor subagent

• /pwdev-code:fleet

  Run multiple roadmap phases end-to-end (plan→execute→review→verify) in parallel, each isolated in its own git worktree + docker-compose stack + tmux pane

• /pwdev-code:gemini

  Delegate a read-only analysis or review task to the Gemini CLI (use --write to allow file changes)

• /pwdev-code:health

  Run a comprehensive project health diagnostic across code, tests, dependencies, security, and process. Use --deps for a focused dependency audit.

• /pwdev-code:init

  Initialize the PWDEV-CODE framework, map existing codebases, configure MCP servers, detect stack, or generate CLAUDE.md

• /pwdev-code:kimi

  Delegate a large-scale exploration or extensive refactor to the Kimi Code CLI, then critically review the resulting diff

• /pwdev-code:kiro

  Delegate an agentic implementation task to the Kiro CLI (headless mode), then critically review the resulting diff

• /pwdev-code:maintenance

  Project maintenance — archive completed artifacts or generate changelog

• /pwdev-code:manager-skills

  Create, list, or audit skills — domain knowledge packs that agents use for quality output

• /pwdev-code:memory

  Manage curated project memory — capture, list, show, and forget durable decisions, lessons, and conventions

• /pwdev-code:opencode

  Delegate a task to the OpenCode CLI (flexible provider/model routing), then critically review the resulting diff

• /pwdev-code:plan

  Run the planning phase to break the spec into waves and executable tasks

• /pwdev-code:product

  Create a PRD, decompose it into an executable roadmap, or generate user stories

• /pwdev-code:quick

  All-in-one quick mode for small tasks (up to 5 files)

• /pwdev-code:review

  Run code review + QA test audit on implemented code. Spawns the code-reviewer and qa subagents in parallel.

• /pwdev-code:session

  Check project status or resume a previous session

• /pwdev-code:simplify

  Analyze the phase diff and apply human-approved high-confidence simplifications via the simplifier subagent

• /pwdev-code:verify

  Run the verification phase to validate implementation against the spec

Subagents (8)

• advisor

  Senior technical advisor: resolves ONE structured decision question raised by a blocked executor (NEEDS_ADVICE) — spec ambiguity, architectural fork, or repeated verification failure. Read-only investigation; picks ONE direction. Dispatched by /pwdev-code:execute. Never implements.

• code-reviewer

  Reviews a diff or file list against spec quality criteria, conventions, and security — reports findings, never fixes code. Dispatched by /pwdev-code:review (in parallel with qa).

• executor

  Implements ONE atomic pwdev-code task: code, verification, atomic commit, and execution summary. Dispatched by /pwdev-code:execute (including fix plans) — do not use for planning, review, or spec work.

• qa

  Audits test coverage and runs the real test suite for implemented tasks; proposes missing-test skeletons. Dispatched by /pwdev-code:review in parallel with code-reviewer. Never writes production code.

• researcher

  Investigates stack, domain, and pitfalls for a feature and writes .planning/context/{domain,stack,pitfalls}.md. Dispatched by /pwdev-code:discover in parallel with the interview. Read-mostly; never makes design decisions.

• roadmap

  Decomposes an approved PRD into the multi-file executable roadmap (.planning/product/roadmap/ hierarchy Phase→Epic→Feature→Task with full traceability). Dispatched by /pwdev-code:product roadmap.

• simplifier

  Analyzes a diff and proposes high-confidence (>=80%) simplifications — reuse of existing code, dead-code removal, complexity reduction, efficiency. Applies ONLY human-approved proposals with its own refactor commit. Dispatched by /pwdev-code:simplify in two passes (ANALYZE, then APPLY). Never hunts

• verifier

  Adversarial verifier: tries to REFUTE that the phase is complete by falsifying each spec truth (objective, ACs, DoD, prohibitions) with real commands. Dispatched by /pwdev-code:verify. Generates fix plans when it rejects.

Hooks (6)

• SessionStart

  "${CLAUDE_PLUGIN_ROOT}/scripts/audit-hook.sh" session_start

• SubagentStart

  "${CLAUDE_PLUGIN_ROOT}/scripts/audit-hook.sh" subagent_start

• SubagentStop

  "${CLAUDE_PLUGIN_ROOT}/scripts/audit-hook.sh" subagent_stop

• PostToolUse

  "${CLAUDE_PLUGIN_ROOT}/scripts/audit-hook.sh" artifact

• Stop

  "${CLAUDE_PLUGIN_ROOT}/scripts/audit-hook.sh" stop

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}/scripts/guard-secrets.sh"

Similar plugins