By tuanldas
Customizable agents, skills, and commands for Claude Code — replaces Superpowers and GSD
Generate API documentation: OpenAPI spec, Postman Collection, and/or Markdown docs from codebase
Brainstorm ideas: explore vague concepts, 2-round research, produce actionable spec
Code review: linting + 1 reviewer agent (13 domains), confidence-scored report, auto-fix
Quick debug: fast fix for simple bugs in main context, auto-escalates to /st:debug if complex
Systematic debugging: persistent state, knowledge base, 12 techniques, anti-shortcut discipline
Analyzes codebase structure and produces structured analysis documents. Spawned by /st:init with a specific focus area (tech, arch, quality, concerns). <example> Context: User runs /st:init on a brownfield project user: "/st:init" assistant: "Brownfield project detected. Spawning codebase-mapper agents to analyze the existing codebase." [mapper receives focus=tech] assistant: "MAPPING COMPLETE — .superteam/mapping/TECH.md" </example>
Investigates bugs using scientific method with persistent session state and knowledge base. Spawned by /st:debug command. <example> Context: User encounters a failing test they can't figure out user: "/st:debug test_auth_flow is failing intermittently" assistant: "Spawning debugger agent to investigate" </example>
Standard implementation specialist. Executes tasks following plans, writes clean code, self-verifies acceptance criteria. Member of /st:team. <example> Context: Scrum Master assigns a task scrum-master: "Task #2: Add theme toggle to settings page. Files: settings.tsx" developer: "Reading settings.tsx, implementing toggle component..." </example>
CI/CD and infrastructure specialist. Handles pipeline configuration, deployment, containerization. Optional role — activated when CI/CD config detected. Member of /st:team. <example> Context: Scrum Master assigns CI/CD task scrum-master: "Task #5: Update GitHub Actions workflow for new feature tests." devops-engineer: "Reading current workflow, adding test step..." </example>
Executes implementation plans with atomic commits, wave-based parallelism, and deviation handling. Spawned by /st:execute and /st:phase-execute commands. <example> Context: A plan has been created and approved for adding dark mode user: "/st:execute" assistant: "Spawning executor agent to implement the plan" </example>
Use when writing code that will be committed. Enforces task-atomic commit granularity, conventional message format, independent revertability, and parallel-safe git operations. Auto-triggered by execute, quick, tdd, phase-execute.
Use when executing any command or agent task. Cross-cutting principles for all Superteam work. Auto-triggered by all commands and agents. Covers: visual-first (design decisions, verification, comparisons, any visual content), UI screenshots, Playwright, browser preview, HTML preview, questioning, one question per message, adaptive, ASK/PRESENT/CONFIRM, recommendation, decision continuity, confirmed decisions, DECISIONS block, never re-ask confirmed.
Use when making design judgments for UI/UX work — auditing, specifying, reviewing, or building frontend interfaces. Covers accessibility, touch targets, typography, color, layout, spacing, animation, forms, navigation, design tokens, and AI slop detection. Reference for agents needing concrete design rules with specific values. Keywords: WCAG, contrast ratio, touch target, font pairing, color palette, responsive breakpoints, design tokens, CSS variables, animation timing, spacing scale, AI slop, anti-patterns.
Use when pausing work mid-session or resuming from a previous session. Defines HANDOFF.json/HANDOFF.md schemas, capture/restore protocols, context quality standards, and one-shot lifecycle rules.
Use when creating any implementation plan (task or phase). Enforces goal-backward plan design, quality gates before execution, anti-vagueness rules, and plan-checker agent verification loop.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Claude Code plugin with customizable agents, skills, and commands tailored to different project types. Auto-detects your project and applies appropriate workflows.
Replaces Superpowers and GSD.
claude plugin add superteam
Or from local directory:
claude plugin add /path/to/superteam
To add Superteam as a marketplace source (enables claude plugin add st@superteam):
# From GitHub
claude /plugin → Add Marketplace → tuanldas/superteam
# From local directory
claude /plugin → Add Marketplace → /path/to/superteam
Requires Node.js >= 18. Zero external dependencies.
All commands use the /st: prefix.
| Command | Description |
|---|---|
/st:init | Initialize project: detect, question, research, requirements, roadmap |
/st:plan | Create implementation plan for a task |
/st:execute | Execute a plan with atomic commits |
/st:quick | Execute a simple task, skip optional steps |
/st:brainstorm | Explore ideas before planning |
| Command | Description |
|---|---|
/st:tdd | Test-driven development workflow |
/st:debug | Systematic debugging with persistent state |
/st:debug-quick | Quick debug, escalates to /st:debug if needed |
/st:code-review | Code review with subagent |
/st:review-feedback | Respond to code review feedback |
/st:worktree | Create isolated git worktree |
| Command | Description |
|---|---|
/st:readme | Generate/update README |
/st:api-docs | Generate API documentation |
/st:ui-design | UI mockup/preview via Playwright MCP |
/st:design-system | Manage design system tokens |
| Command | Description |
|---|---|
/st:phase-add | Add phase to roadmap |
/st:phase-remove | Remove phase from roadmap |
/st:phase-list | List all phases |
/st:phase-discuss | Discuss context before planning |
/st:phase-research | Deep research with parallel agents |
/st:phase-plan | Create detailed phase plan |
/st:phase-execute | Execute phase with wave parallelism |
/st:phase-validate | Validate phase completion |
| Command | Description |
|---|---|
/st:milestone-new | Create new milestone |
/st:milestone-audit | Audit milestone completion |
/st:milestone-complete | Complete and archive milestone |
/st:milestone-archive | Clean up phase files |
| Command | Description |
|---|---|
/st:team create | Create Scrum team of specialized agents |
/st:team run | SM orchestrates roadmap phases with checkpoints |
/st:team status | Show team dashboard |
/st:team disband | Graceful team shutdown |
| Command | Description |
|---|---|
/st:pause | Save context for later (handoff files + WIP commit) |
/st:resume | Restore context from previous session |
Skills are cross-cutting methodologies shared across commands. They enforce discipline, not just provide guidance.
| Skill | Used by |
|---|---|
core-principles | All commands and agents (auto-injected) |
project-awareness | All commands (via session-start) |
atomic-commits | execute, quick, tdd, phase-execute |
research-methodology | init, phase-research, brainstorm |
scientific-debugging | debug, debug-quick |
tdd-discipline | tdd |
requesting-code-review | code-review |
receiving-code-review | review-feedback |
wave-parallelism | execute, phase-execute |
handoff-protocol | pause, resume |
plan-quality | plan, phase-plan |
verification | phase-validate, milestone-audit |
frontend-design | ui-design, design-system, ui-auditor, ux-designer |
team-coordination | team, developer, senior-developer (team context) |
22 specialized subagents for parallel and focused work:
reviewer planner executor debugger verifier phase-researcher research-orchestrator research-synthesizer roadmapper codebase-mapper integration-checker plan-checker ui-researcher ui-auditor ux-designer test-auditor developer senior-developer tech-lead scrum-master qa-engineer devops-engineer
Superteam auto-detects your project type on session start:
| Detection | Markers |
|---|---|
| Frontend | react, vue, @angular/core, svelte |
| Backend | express, fastify, @nestjs/core, koa, hono |
| Fullstack | next, nuxt, remix |
| PHP | composer.json, artisan |
| Go | go.mod |
| Python | pyproject.toml, requirements.txt |
| Rust | Cargo.toml |
| Monorepo | package.json workspaces, pnpm-workspace.yaml, lerna.json |
After /st:init, config lives at .superteam/config.json:
npx claudepluginhub tuanldas/superteam --plugin stStructured project planning and execution through brainstorm, spec, and build phases across three execution tiers: sequential, delegated sub-agents, and full agent teams
Agent Teams orchestration, governance hooks, multi-AI review, memento skill intelligence, and project management skills for Claude Code
Multi-agent team orchestration for parallel task execution, research, and implementation
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development