Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By Goktug
Orchestrates an AI crew that takes any React Native or TypeScript backend task from idea to pull request with a single human checkpoint — a team-lead agent dispatches specialist developers, QA engineers, security auditors, and web researchers to handle specification, planning, implementation, testing, review, and deployment readiness.
npx claudepluginhub goktug/ai-crew --plugin ai-crewImplement the next task incrementally — build, test, verify, commit
Simplify code for clarity and maintainability — reduce complexity without changing behavior
Break work into small verifiable tasks with acceptance criteria and dependency ordering
Conduct a five-axis code review — correctness, readability, architecture, security, performance
Run the pre-launch checklist and prepare for production deployment
Senior code reviewer that evaluates changes across five dimensions — correctness, readability, architecture, security, and performance. Use for thorough code review before merge.
Opus-backed implementation engineer for complex tasks requiring deeper reasoning (complex complexity). Executes one task per dispatch from an ai-crew team-lead. Reads the exact spec and plan line ranges cited in the prompt plus the required skills; follows TDD and incremental-implementation; returns a one-line PASS or FAIL summary. Cannot dispatch subagents — strict 1-level dispatch.
Security engineer focused on vulnerability detection, threat modeling, and secure coding practices. Use for security-focused code review, threat analysis, or hardening recommendations.
Sonnet-backed implementation engineer for atomized, mechanical tasks (simple complexity). Executes one task per dispatch from an ai-crew team-lead. Reads the exact spec and plan line ranges cited in the prompt plus the required skills; follows TDD and incremental-implementation; returns a one-line PASS or FAIL summary. Cannot dispatch subagents — strict 1-level dispatch.
QA engineer specialized in test strategy, test writing, and coverage analysis. Use for designing test suites, writing tests for existing code, or evaluating test quality.
Guides stable API and interface design. Use when designing APIs, module boundaries, or any public interface. Use when creating REST or GraphQL endpoints, defining type contracts between modules, or establishing boundaries between frontend and backend.
Tests in real browsers. Use when building or debugging anything that runs in a browser. Use when you need to inspect the DOM, capture console errors, analyze network requests, profile performance, or verify visual output with real runtime data via Chrome DevTools MCP.
Automates CI/CD pipeline setup. Use when setting up or modifying build and deployment pipelines. Use when you need to automate quality gates, configure test runners in CI, or establish deployment strategies.
Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.
Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Complete developer toolkit for Claude Code
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams
A battle-tested, production-grade engineering orchestrator for Claude Code.
Describe the job. Approve the plan. Get the PR.
ai-crew encodes the workflows, quality gates, and best practices senior engineers run on every change — intake, spec, plan, build, verify, review, ship — and packages them as a Claude Code orchestrator for TypeScript, Node, backend, and React Native. An Opus team-lead reasons through every phase; scoped Sonnet developers handle implementation in sequence or in parallel; one human checkpoint sits between Plan and Build.
Built on top of Addy Osmani's agent-skills (vendored, MIT). See § Why this is different for what ai-crew adds on top.
I love Superpowers and Addy Osmani's
agent-skills, but I kept finding myself too much in the loop. All I wanted was to define the spec, approve the plan, and get a tested, working PR — not narrate the next step or remember which skill to invoke.So I started with the obvious AI-team setup: architect, business analyst, QA lead, researcher, designer, developer, each as its own subagent. Burned tokens. Quality barely changed. So I cut them.
The unlock: main context stays lean. The orchestrator does analysis, spec, plan, and review inline. Only coding tasks and one-shot web research get delegated — the parts that benefit from a fresh, narrow context. Developers run in parallel when tasks are independent, in sequence when they're not.
That's ai-crew's team-lead. One brain at the wheel, scoping every task and delegating only what needs a fresh, narrow context. Guided Q&A for spec, and one human checkpoint for plan approval.
$ claude
> /team-lead Wire RevenueCat into our paywall, gate the premium tab, and track
checkout events with our analytics SDK
ai-crew [Intake — Q1 of 3] one question at a time, A/B/C with tradeoffs
How should the premium gate behave when a user without entitlement
taps the tab?
A. Hard gate — replace the tab entirely with the paywall
(recommended: cleanest mental model, zero leakage of premium UI)
B. Modal paywall over the existing tab; dismiss returns to free content
(best if you have free content worth keeping inside the tab)
C. Soft gate — inline teaser with a Subscribe CTA
(highest conversion in some apps, mixed UX consistency)
You B
ai-crew [Q2 — RevenueCat SDK version?] A: 4.x B: 5.0 C: other
You A
ai-crew [Q3 — Analytics destination?] A: Segment B: Amplitude C: custom
You C — Mixpanel via our own wrapper
ai-crew [Intake complete. Writing spec...]
[spec.md written]
[plan.md written — 6 tasks, 2 parallel waves]
✋ CHECKPOINT — review plan.md before I write a single file.
Approve, or tell me what to change.
You Looks good, go.
ai-crew [Wave 1 — dispatching 3 parallel Sonnet developers...]
[Wave 2 — dispatching 2 parallel Sonnet developers...]
[Verify — all tests green]
[Review — inline five-axis review, no critical findings]
[Ship — PR opened]
PR: https://github.com/your-org/your-app/pull/47
"feat: RevenueCat paywall, premium gate, checkout analytics"
spec.md and plan.md before any code is touched — edit them if you want.Click any tool to expand. Claude Code is the primary platform; the skills run anywhere agent-skills do.
Marketplace install:
/plugin marketplace add goktug/ai-crew
/plugin install ai-crew@ai-crew
SSH errors? The marketplace clones repos via SSH. If you don't have SSH keys set up on GitHub, either add an SSH key or switch to HTTPS for fetches:
git config --global url."https://github.com/".insteadOf "git@github.com:"
Local / development:
git clone https://github.com/goktug/ai-crew.git
claude --plugin-dir /path/to/ai-crew
Try it:
> /team-lead Migrate our Express auth middleware to JWT rotation, with tests and a rollback plan.