claude

dev-team

An AI-powered product development team covering discovery, UX, architecture, implementation, code review, and QA — with human approval gates at every critical decision. Pilot project is a Notes/Reminders Android app.

0 on GitHubby David Lunav0.1.0MITupdated 3 months agoworkflow
/

Plugin dev-team from marketplace ai-product-dev-team-marketplace in piroworkz/AIProductDevTeam. Commands read from the repo's manifest — 1 month 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.

What it installs

Skills (8)

  • approve-architecture

    Approve Gate 3 (Architecture Review) for a story. Advances state from ARCHITECTURE_REVIEW to TECHNICAL_PLANNING and invokes the tech-lead agent to produce the implementation task breakdown.

  • approve-requirements

    Approve Gate 1 (Requirements Review) for a story. Advances state from REQUIREMENTS_REVIEW to UX_DESIGN and invokes the ux-agent to begin the UX design phase.

  • approve-ux

    Approve Gate 2 (UX Review) for a story. Advances state from UX_REVIEW to ARCHITECTURE and invokes the architect agent to begin the architecture phase.

  • implement-android

    Run the Android implementation phase for a story. Invokes android-dev in an isolated work tree, then tech-lead for code review, and stops at the QA gate for the qa-agent to validate acceptance criteria.

  • implement-ios

    Run the iOS implementation phase for a story. Invokes ios-dev in an isolated work tree, then tech-lead for code review, and stops at the QA gate for the qa-agent to validate acceptance criteria.

  • merge-story

    Approve Gate 4 (Merge Approval) for a story that has passed QA. Advances state to DONE, marks all artifacts as approved, and confirms the story is complete.

  • qa-gate

    Run the QA gate for a story that has completed code review. Invokes the qa-agent to build, test, and validate behavior, produces a QA report, and surfaces Gate 4 (Merge Approval) to the human.

  • start-feature

    Start a new feature through the product discovery phase. Invokes the po-agent to produce a problem statement, user story, and acceptance criteria, then stops at Gate 1 (Requirements Review) for human review.

Subagents (9)

  • android-dev

    Android Developer — implements Kotlin/Compose features from approved technical plans. Operates in IMPLEMENTATION state. Works in an isolated Git worktree. Never modifies product, design, or architecture documents.

  • architect

    Software Architect — proposes module structure, data contracts, and ADRs for approved UX specs. Operates in ARCHITECTURE state. Output stops at the Architecture gate for human review.

  • ios-dev

    iOS Developer — implements Swift/SwiftUI features from approved technical plans. Operates in IMPLEMENTATION state. Works in an isolated Git worktree. Currently inactive — iOS platform is future scope.

  • orchestrator

    State machine router — reads the current workflow state from project.yaml, delegates to the correct agent, and surfaces approval gates to the human owner. The only agent that may spawn other agents.

  • po-agent

    Product Owner — translates ideas into problem statements, user stories, and acceptance criteria. Operates in PRODUCT_DISCOVERY state. Output stops at the Requirements gate for human review.

  • qa-agent

    QA Engineer — validates implementation against acceptance criteria, runs build and test commands, and produces a release recommendation. Operates in QA state. Never modifies implementation code.

  • sm-agent

    Scrum Master — tracks workflow state across all active stories, detects blockers, maintains the backlog status, and surfaces impediments to the orchestrator. Does not write product or technical artifacts.

  • tech-lead

    Tech Lead — decomposes approved architecture into implementation tasks, writes the technical plan, and performs code review. Operates in TECHNICAL_PLANNING and CODE_REVIEW states.

  • ux-agent

    UX Designer — translates approved acceptance criteria into user flows and screen specs. Operates in UX_DESIGN state. Output stops at the UX gate for human review.

Hooks (3)

  • PreToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/block-secret-reads.sh

  • PreToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/block-destructive-git.sh

  • PostToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/log-changed-files.sh

View source on GitHub

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

Backend API design, GraphQL architecture, workflow orchestration with Temporal, and test-driven backend development

9 skills · 1 command · 8 agents

wshobson/agents 42,461

beads

claudecodex

AI-supervised issue tracker for coding workflows

1 skill · 1 agent · hooks

gastownhall/beads 28,698

superset

claude

Use Superset to spawn agents, create workspaces, manage tasks, and schedule automations from Claude Code.

14 skills · 1 command

superset-sh/superset 15,548

MiniMax AI skills library for frontend, fullstack, Android, iOS, shader, GIF sticker, document, presentation, spreadsheet, and multimodal media workflows

18 skills

MiniMax-AI/skills 13,828

worktrunk

claudecodex

Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. This plugin provides configuration guidance (LLM commit messages, project hooks, worktree paths), automatic activity tracking (🤖/💬 indicators in `wt list` showing active Claude sessions), and a `/wt-diff` pane showing each worktree's changes.

2 skills · hooks

max-sixty/worktrunk 12,481

Maintain dev-team?

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.

claude & codex pluginslisted