team-composition-patterns
starsforks
How to install team-composition-patterns?
wshobson/agents/team-composition-patterns
$npx -y skills add wshobson/agents --skill team-composition-patternsDesign 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 corr
SKILL.md
| 1 | # Team Composition Patterns |
| 2 | |
| 3 | Best practices for composing multi-agent teams, selecting team sizes, choosing agent types, and configuring display modes for Claude Code's Agent Teams feature. |
| 4 | |
| 5 | ## When to Use This Skill |
| 6 | |
| 7 | - Deciding how many teammates to spawn for a task |
| 8 | - Choosing between preset team configurations |
| 9 | - Selecting the right agent type (subagent_type) for each role |
| 10 | - Configuring teammate display modes (tmux, iTerm2, in-process) |
| 11 | - Building custom team compositions for non-standard workflows |
| 12 | |
| 13 | ## Team Sizing Heuristics |
| 14 | |
| 15 | | Complexity | Team Size | When to Use | |
| 16 | | ------------ | --------- | ----------------------------------------------------------- | |
| 17 | | Simple | 1-2 | Single-dimension review, isolated bug, small feature | |
| 18 | | Moderate | 2-3 | Multi-file changes, 2-3 concerns, medium features | |
| 19 | | Complex | 3-4 | Cross-cutting concerns, large features, deep debugging | |
| 20 | | Very Complex | 4-5 | Full-stack features, comprehensive reviews, systemic issues | |
| 21 | |
| 22 | **Rule of thumb**: Start with the smallest team that covers all required dimensions. Adding teammates increases coordination overhead. |
| 23 | |
| 24 | ## Preset Team Compositions |
| 25 | |
| 26 | ### Review Team |
| 27 | |
| 28 | - **Size**: 3 reviewers |
| 29 | - **Agents**: 3x `team-reviewer` |
| 30 | - **Default dimensions**: security, performance, architecture |
| 31 | - **Use when**: Code changes need multi-dimensional quality assessment |
| 32 | |
| 33 | ### Debug Team |
| 34 | |
| 35 | - **Size**: 3 investigators |
| 36 | - **Agents**: 3x `team-debugger` |
| 37 | - **Default hypotheses**: 3 competing hypotheses |
| 38 | - **Use when**: Bug has multiple plausible root causes |
| 39 | |
| 40 | ### Feature Team |
| 41 | |
| 42 | - **Size**: 3 (1 lead + 2 implementers) |
| 43 | - **Agents**: 1x `team-lead` + 2x `team-implementer` |
| 44 | - **Use when**: Feature can be decomposed into parallel work streams |
| 45 | |
| 46 | ### Fullstack Team |
| 47 | |
| 48 | - **Size**: 4 (1 lead + 3 implementers) |
| 49 | - **Agents**: 1x `team-lead` + 1x frontend `team-implementer` + 1x backend `team-implementer` + 1x test `team-implementer` |
| 50 | - **Use when**: Feature spans frontend, backend, and test layers |
| 51 | |
| 52 | ### Research Team |
| 53 | |
| 54 | - **Size**: 3 researchers |
| 55 | - **Agents**: 3x `general-purpose` |
| 56 | - **Default areas**: Each assigned a different research question, module, or topic |
| 57 | - **Capabilities**: Codebase search (Grep, Glob, Read), web search (WebSearch, WebFetch) |
| 58 | - **Use when**: Need to understand a codebase, research libraries, compare approaches, or gather information from code and web sources in parallel |
| 59 | |
| 60 | ### Security Team |
| 61 | |
| 62 | - **Size**: 4 reviewers |
| 63 | - **Agents**: 4x `team-reviewer` |
| 64 | - **Default dimensions**: OWASP/vulnerabilities, auth/access control, dependencies/supply chain, secrets/configuration |
| 65 | - **Use when**: Comprehensive security audit covering multiple attack surfaces |
| 66 | |
| 67 | ### Migration Team |
| 68 | |
| 69 | - **Size**: 4 (1 lead + 2 implementers + 1 reviewer) |
| 70 | - **Agents**: 1x `team-lead` + 2x `team-implementer` + 1x `team-reviewer` |
| 71 | - **Use when**: Large codebase migration (framework upgrade, language port, API version bump) requiring parallel work with correctness verification |
| 72 | |
| 73 | ## Agent Type Selection |
| 74 | |
| 75 | When spawning teammates with the `Agent` tool, choose `subagent_type` based on what tools the teammate needs: |
| 76 | |
| 77 | | Agent Type | Tools Available | Use For | |
| 78 | | ------------------------------ | ---------------------------------------------------- | ---------------------------------------------------------- | |
| 79 | | `general-purpose` | All tools (Read, Write, Edit, Bash, etc.) | Implementation, debugging, any task requiring file changes | |
| 80 | | `Explore` | Read-only tools (Read, Grep, Glob) | Research, code exploration, analysis | |
| 81 | | `Plan` | Read-only tools | Architecture planning, task decomposition | |
| 82 | | `agent-teams:team-reviewer` | Read/search/Bash plus TaskList/TaskGet/TaskUpdate/SendMessage | Code review with structured findings | |
| 83 | | `agent-teams:team-debugger` | Read/search/Bash plus TaskList/TaskGet/TaskUpdate/SendMessage | Hypothesis-driven investigation | |
| 84 | | `agent-teams:team-implementer` | Read/Write/Edit/search/Bash plus TaskList/TaskGet/TaskUpdate/SendMessage | Building features within file ownersh |
People who liked this also viewed
6 picksType
- find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 2026
- grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026
- vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026
- improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026
- setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026
- triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026
