team — Claude Code plugin: what it does, how to install
team — a coordinated set of specialist agents for software delivery, covering requirements, architecture, coding, testing, UX, and operations.
MIT ★ 587 updated 2026-10-26
/plugin marketplace add rsmdt/the-startup
/plugin install team@the-startupWhat team does
The plugin brings together 8 roles, 15 activity-focused agents, and 14 reusable skills for software work. Its roles cover product analysis, system architecture, implementation, testing, design, platform operations, task routing, and agent creation.
Agents decide who handles an activity and when; skills describe the reusable methods behind that work. The library includes product research, API and system design, feature building, performance analysis, security review, test planning, user research, and production monitoring. It runs in Claude Code. Since activities are split by specialist role, choosing the matching activity first is the practical way to use the team.
What it helps with
- clarify product requirements and priorities
- design service boundaries and data strategy
- build features across UI, API, and data layers
- plan risk-based tests and load validation
- set up delivery platforms and production monitoring
Example prompts for the agent
- Research this product, resolve conflicting requirements, and define acceptance criteria.
- Design the system architecture for this feature, including service boundaries and deployment strategy.
- Create a test strategy for this project with risk coverage and load validation.
How to install team
Plugin marketplace
/plugin marketplace add rsmdt/the-startup/plugin install team@the-startupStep by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own
Questions about team
How do I install team in Claude Code?
Run /plugin marketplace add rsmdt/the-startup and then /plugin install team@the-startup in Claude Code. The commands are /plugin marketplace add rsmdt/the-startup followed by /plugin install team@the-startup.
What does the team plugin do?
It adds specialist agents and reusable skills for software delivery. The agents cover requirements, architecture, implementation, testing, UX, platform operations, and agent design.
What is included in team?
The plugin contains 8 roles, 15 activity-based agents, and 14 skills. Agents can reference skills through YAML frontmatter, such as project discovery, API contract design, testing, and security assessment.
Source text
team · MIT
# Team Plugin - The Agentic Startup **Specialized agent library for Claude Code with consolidated, high-signal activities for software delivery.** The `team` plugin provides **8 specialized roles**, **15 activity-based agents**, and **14 reusable skills**. This consolidation keeps specialist depth while reducing overlap and context bloat. --- ## Installation ```bash /plugin install team@the-startup ``` --- ## Agent Roles ### The Chief **Complexity assessment and activity routing specialist** - `the-chief` ### The Analyst (1 activity) **Product research and requirements specialist** | Activity | Focus | |----------|-------| | `research-product` | Market analysis, requirement clarification, prioritization, stakeholder alignment | ### The Architect (4 activities) **System design and technical governance specialist** | Activity | Focus | |----------|-------| | `design-system` | Scalable architecture, service boundaries, data + deployment strategy | | `review-security` | Application security + dependency/supply-chain review | | `review-robustness` | Complexity and concurrency risk review | | `review-compatibility` | Breaking-change detection, migration safety, backwards compatibility | ### The Developer (2 activities) **Implementation and optimization specialist** | Activity | Focus | |----------|-------| | `build-feature` | Feature implementation across UI, API, services, and data layers | | `optimize-performance` | Bottleneck diagnosis and targeted optimization | ### The Tester (1 activity) **Functional and performance testing specialist** | Activity | Focus | |----------|-------| | `test-strategy` | Risk-based test planning, quality coverage, load/stress validation | ### The Designer (3 activities) **User research and UX design specialist** | Activity | Focus | |----------|-------| | `research-user` | User interviews, persona/journey insights, behavioral evidence | | `design-interaction` | Information architecture, user flows, interaction patterns | | `design-visual` | Design systems plus accessibility-by-default standards | ### The DevOps (2 activities) **Platform and operations specialist** | Activity | Focus | |----------|-------| | `build-platform` | Containers + IaC + CI/CD as one delivery platform | | `monitor-production` | Observability, SLI/SLOs, alerting, incident diagnostics | ### The Meta Agent **Agent design and generation specialist** - `the-meta-agent` --- ## Skills System Skills are referenced in agent YAML frontmatter: ```yaml --- name: agent-name skills: project-discovery, pattern-detection, api-contract-design --- ``` ### Available Skills (14) | Category | Skill | Description | |----------|-------|-------------| | **Cross-Cutting** | `project-discovery` | Unified navigation, stack detection, and doc validation | | | `pattern-detection` | Identify existing codebase patterns for consistency | | | `feature-prioritization` | RICE/MoSCoW/Kano/value-effort prioritization | | | `requirements-elicitation` | Requirement gathering, conflict resolution, acceptance criteria | | **Design** | `user-research` | Research planning, evidence synthesis, personas, journey mapping | | **Development** | `api-contract-design` | REST/GraphQL contract design and versioning | | | `architecture-selection` | Pattern selection for monolith/microservices/serverless/event-driven | | | `domain-modeling` | Domain + data modeling, invariants, schema evolution | | | `technical-writing` | ADRs, system docs, runbooks, API docs | | | `testing` | Layered testing strategy and execution patterns | | **Infrastructure** | `platform-operations` | Pipeline design + observability + release reliability controls | | **Quality** | `code-quality-review` | Holistic review including security/perf/a11y/error handling | | | `performance-analysis` | Profiling and bottleneck identification | | | `security-assessment` | Threat modeling and vulnerability review | --- ## Architecture Principles 1. Agents define **who/what/when**. 2. Skills provide reusable **how**. 3. Consolidation favors clearer boundaries and lower context overhead. 4. Deliverables stay concrete and execution-oriented.
© Rudolf S. · MIT · rsmdt/the-startup