team-composition-patterns
wshobson/agents
Design optimal multi-agent team compositions with sizing heuristics, preset configurations, and agent type selection.
What is team-composition-patterns?
This skill provides best practices for composing Claude Code Agent Teams, including team sizing heuristics (1–5 agents based on complexity), preset configurations for common workflows (review, debug, feature, fullstack, research, security, migration teams), and guidance on selecting the right agent type for each role. Use it when deciding team size, choosing between preset team patterns, selecting subagent_type for specialized roles, configuring display modes, or building custom team compositions.
- Apply team sizing heuristics based on task complexity (simple 1–2 agents, moderate 2–3, complex 3–4, very complex 4–5)
- Select from seven preset team compositions: Review, Debug, Feature, Fullstack, Research, Security, and Migration teams
- Match agent types to required capabilities: general-purpose, Explore, Plan, team-reviewer, team-debugger, team-implementer, team-lead
- Configure teammate display modes (tmux, iTerm2, in-process) for different environments
- Define clear role boundaries and file ownership to minimize coordination overhead
- Troubleshoot common team composition issues: read-only agent limitations, coordination overhead, overlapping reviewer coverage
How to install team-composition-patterns
npx skills add https://github.com/wshobson/agents --skill team-composition-patternsHow to use team-composition-patterns
- 1.Assess task complexity (simple, moderate, complex, very complex) to determine team size
- 2.Select a preset team composition matching your workflow (review, debug, feature, fullstack, research, security, migration) or design a custom team
- 3.Choose the correct subagent_type for each role based on required tools (general-purpose for implementation, Explore/Plan for read-only research, specialized agents for review/debug/implementation)
- 4.Configure display mode in ~/.claude/settings.json (tmux for development, in-process for CI/CD)
- 5.Define clear ownership boundaries and responsibilities for each teammate before spawning
- 6.Spawn teammates using the Agent tool with complete context in initial prompts, as they start with no prior conversation history
Use cases
- Deciding whether to spawn 2 or 4 agents for a multi-file feature implementation
- Choosing a Review Team (3 reviewers) for security and architecture assessment across code changes
- Assembling a Fullstack Team (1 lead + frontend/backend/test implementers) for end-to-end feature development
- Building a Security Team (4 reviewers) for comprehensive audit covering OWASP, auth, dependencies, and secrets
- Composing a Migration Team (1 lead + 2 implementers + 1 reviewer) for large codebase upgrades with parallel work and verification
- Engineering leads and architects designing multi-agent workflows
- Developers implementing features that benefit from parallel work streams
- Code reviewers and security teams conducting comprehensive audits
- DevOps and CI/CD engineers configuring agent teams in automated environments
- Teams performing large migrations or systemic refactors
team-composition-patterns FAQ
Use the team sizing heuristic: 1–2 for simple tasks (single-dimension review, isolated bugs), 2–3 for moderate complexity (multi-file changes, 2–3 concerns), 3–4 for complex work (cross-cutting concerns, large features), and 4–5 for very complex tasks (full-stack features, systemic issues). Start with the smallest team covering all required dimensions; each additional teammate increases coordination overhead.
general-purpose agents have all tools (Read, Write, Edit, Bash) and can implement features. Explore and Plan are read-only (Grep, Glob, Read only) for research and analysis. Specialized agents (team-reviewer, team-debugger, team-implementer, team-lead) have targeted tools plus team coordination capabilities. Never assign implementation tasks to read-only agents.
Use preset compositions (Review, Debug, Feature, Fullstack, Research, Security, Migration) when your workflow matches the pattern. Build custom teams when you have non-standard requirements. Always ensure every custom team has a coordinator (team-lead or user), matches roles to agent types, avoids duplicate roles, defines clear ownership boundaries, and keeps size to 2–4 agents.
Use tmux mode for development workflows where you want to monitor multiple agents in separate panes. Use iTerm2 mode on macOS if you prefer iTerm2 tabs. Use in-process mode for CI/CD environments or simple tasks where tmux is unavailable. Configure in ~/.claude/settings.json.
Review dimensions are overlapping. Redefine each reviewer's focus area explicitly: one on correctness/logic, one on security, one on performance/scalability. This prevents duplicate findings and wasted tokens.
Full instructions (SKILL.md)
Source of truth, from wshobson/agents.
name: team-composition-patterns description: Design optimal agent team compositions with sizing heuristics, preset configurations, and agent type selection. Use this skill when deciding how many agents to spawn for a task, when choosing between a review team versus a feature team versus a debug team, when selecting the correct subagent_type for each role to ensure agents have the tools they need, when configuring display modes (tmux, iTerm2, in-process) for a CI or local environment, or when building a custom team composition for a non-standard workflow such as a migration or security audit. version: 1.0.2
Team Composition Patterns
Best practices for composing multi-agent teams, selecting team sizes, choosing agent types, and configuring display modes for Claude Code's Agent Teams feature.
When to Use This Skill
- Deciding how many teammates to spawn for a task
- Choosing between preset team configurations
- Selecting the right agent type (subagent_type) for each role
- Configuring teammate display modes (tmux, iTerm2, in-process)
- Building custom team compositions for non-standard workflows
Team Sizing Heuristics
| Complexity | Team Size | When to Use |
|---|---|---|
| Simple | 1-2 | Single-dimension review, isolated bug, small feature |
| Moderate | 2-3 | Multi-file changes, 2-3 concerns, medium features |
| Complex | 3-4 | Cross-cutting concerns, large features, deep debugging |
| Very Complex | 4-5 | Full-stack features, comprehensive reviews, systemic issues |
Rule of thumb: Start with the smallest team that covers all required dimensions. Adding teammates increases coordination overhead.
Preset Team Compositions
Review Team
- Size: 3 reviewers
- Agents: 3x
team-reviewer - Default dimensions: security, performance, architecture
- Use when: Code changes need multi-dimensional quality assessment
Debug Team
- Size: 3 investigators
- Agents: 3x
team-debugger - Default hypotheses: 3 competing hypotheses
- Use when: Bug has multiple plausible root causes
Feature Team
- Size: 3 (1 lead + 2 implementers)
- Agents: 1x
team-lead+ 2xteam-implementer - Use when: Feature can be decomposed into parallel work streams
Fullstack Team
- Size: 4 (1 lead + 3 implementers)
- Agents: 1x
team-lead+ 1x frontendteam-implementer+ 1x backendteam-implementer+ 1x testteam-implementer - Use when: Feature spans frontend, backend, and test layers
Research Team
- Size: 3 researchers
- Agents: 3x
general-purpose - Default areas: Each assigned a different research question, module, or topic
- Capabilities: Codebase search (Grep, Glob, Read), web search (WebSearch, WebFetch)
- Use when: Need to understand a codebase, research libraries, compare approaches, or gather information from code and web sources in parallel
Security Team
- Size: 4 reviewers
- Agents: 4x
team-reviewer - Default dimensions: OWASP/vulnerabilities, auth/access control, dependencies/supply chain, secrets/configuration
- Use when: Comprehensive security audit covering multiple attack surfaces
Migration Team
- Size: 4 (1 lead + 2 implementers + 1 reviewer)
- Agents: 1x
team-lead+ 2xteam-implementer+ 1xteam-reviewer - Use when: Large codebase migration (framework upgrade, language port, API version bump) requiring parallel work with correctness verification
Agent Type Selection
When spawning teammates with the Agent tool, choose subagent_type based on what tools the teammate needs:
| Agent Type | Tools Available | Use For |
|---|---|---|
general-purpose | All tools (Read, Write, Edit, Bash, etc.) | Implementation, debugging, any task requiring file changes |
Explore | Read-only tools (Read, Grep, Glob) | Research, code exploration, analysis |
Plan | Read-only tools | Architecture planning, task decomposition |
agent-teams:team-reviewer | Read/search/Bash plus TaskList/TaskGet/TaskUpdate/SendMessage | Code review with structured findings |
agent-teams:team-debugger | Read/search/Bash plus TaskList/TaskGet/TaskUpdate/SendMessage | Hypothesis-driven investigation |
agent-teams:team-implementer | Read/Write/Edit/search/Bash plus TaskList/TaskGet/TaskUpdate/SendMessage | Building features within file ownership boundaries |
agent-teams:team-lead | Read/search/Bash plus Agent Teams coordination tools | Team orchestration and coordination |
Key distinction: Read-only agents (Explore, Plan) cannot modify files. Never assign implementation tasks to read-only agents.
Display Mode Configuration
Configure in ~/.claude/settings.json:
{
"teammateMode": "tmux"
}
| Mode | Behavior | Best For |
|---|---|---|
"tmux" | Each teammate in a tmux pane | Development workflows, monitoring multiple agents |
"iterm2" | Each teammate in an iTerm2 tab | macOS users who prefer iTerm2 |
"in-process" | All teammates in same process | Simple tasks, CI/CD environments |
Custom Team Guidelines
When building custom teams:
- Every team needs a coordinator — Either designate a
team-leador have the user coordinate directly - Match roles to agent types — Use specialized agents (reviewer, debugger, implementer) when available
- Avoid duplicate roles — Two agents doing the same thing wastes resources
- Define boundaries upfront — Each teammate needs clear ownership of files or responsibilities
- Keep it small — 2-4 teammates is the sweet spot; 5+ requires significant coordination overhead
Troubleshooting
A teammate was spawned as Explore but needs to write files.
Explore and Plan are read-only agents. Change the subagent_type to general-purpose or an appropriate specialized agent type. Never assign implementation tasks to read-only agents.
The team is growing too large and coordination is slowing everything down. Each additional teammate adds communication overhead. Consolidate roles: can one agent cover two dimensions? A 4-person team doing 6 independent tasks is usually better served by 3 agents covering 2 tasks each.
tmux mode is not showing panes.
Ensure tmux is installed and a session is already running before spawning teammates. The in-process mode works without tmux and is suitable for CI or scripted environments.
Two reviewers are flagging the same issues. The review dimensions overlap. Redefine each reviewer's focus area: one on correctness/logic, one on security, one on performance/scalability. Overlapping coverage wastes tokens and produces duplicate findings.
A team-lead is spawning teammates but they are not receiving tasks.
Verify that the lead is using the Agent tool to spawn teammates and passing complete context in the prompt. Teammates start fresh with no prior conversation history — they need all relevant information in their initial prompt.
Related Skills
- parallel-feature-development — Decompose work streams and assign file ownership once the team is composed
- team-communication-protocols — Establish messaging norms and shutdown procedures for the assembled team
Related skills
More from wshobson/agents and the wider catalog.

temporal-python-testing
Test Temporal workflows with pytest, time-skipping, and mocking strategies.

terraform-module-library
Build reusable Terraform modules for AWS, Azure, GCP, and OCI following infrastructure-as-code best practices.

threat-mitigation-mapping
Map threats to security controls and mitigations for effective defense planning.

trace-to-training-data
Convert graded evaluation traces into SFT examples and DPO preference pairs for model training.

track-management
Create and manage Conductor tracks—logical work units for features, bugs, and refactors.

turborepo-caching
Configure Turborepo for efficient monorepo builds with local and remote caching.