Skip to main content
skill CX Codex

agent-teams · travisjneuman — Claude Code skill: what it does, how to install

Teaches Claude how to compose agent teams: which tasks split well, how many teammates to spawn, and what goes into each spawn prompt. Active only once teams are enabled.

MIT 153 installs ★ 114 updated 2026-10-25

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add travisjneuman/.claude --skill agent-teams -a claude-code
Open on GitHub

What agent-teams does

The skill gives the model a reference on composing Claude Agent Teams. It lists the tasks worth splitting: a broad code review, work that crosses several layers, a bug with no obvious cause, a comparison of approaches. It also lists the ones that are not, such as same-file edits or a tight token budget.

The skill describes six team templates. Review puts three teammates on security, performance and correctness. Feature work runs an architect who plans, an implementer who writes, and a reviewer who checks the diff. Debugging spreads 3-5 competing hypotheses and has teammates argue them out. Two more layouts cover cross-platform apps and full-stack work with per-layer file ownership. Every spawn prompt follows a five-part formula: role, directories, focus, context, deliverable.

Teams stay off until a session starts with CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1. Without it the skill points to ordinary subagents and tells the model not to suggest a team on its own. Platform support differs: macOS and Linux get in-process plus tmux split panes, Windows only in-process, and --teammate-mode auto picks the mode.

What it helps with

  • Decide whether a task needs a team or a single subagent
  • Run a code review from three independent angles
  • Split a full-stack task by file ownership
  • Write a spawn prompt that keeps a teammate inside its directories
  • Test competing theories about a bug in parallel

Example prompts for the agent

  • Put together a three-reviewer team: security, performance, correctness.
  • The bug's cause is unclear — give four hypotheses to separate teammates and let them argue.
  • Lay out spawn prompts for a full-stack task: frontend, backend, database, each with its own directories.

How to install agent-teams

1Skills CLI (Vercel)

npx skills add travisjneuman/.claude --skill agent-teams -a claude-code

2Manually: put the files in place

~/.claude/skills/agent-teams/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/agent-teams/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about agent-teams

How do I install agent-teams in Claude Code?

Run npx skills add travisjneuman/.claude --skill agent-teams -a claude-code; the skill lands at ~/.claude/skills/agent-teams/SKILL.md.

What does agent-teams do?

It gives the model guidance on composing Claude Agent Teams: which tasks fit, how many teammates to spawn, and how to phrase each spawn prompt.

Do I need to enable anything first?

Yes. Start the session with CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1; otherwise the skill falls back to ordinary subagents.

Source text

SKILL.md · MIT
---
name: agent-teams
description: Team composition knowledge for Claude Code Agent Teams - when teams fit, optimal sizing, spawn prompt patterns. Use only when the user has explicitly enabled agent teams for the session.
allowed-tools: []
---

# Agent Teams Composition Skill

You have expertise in composing and orchestrating Claude Code Agent Teams. Use this knowledge when the user's task would benefit from parallel multi-session work.

Agent teams are off by default. Use them only when the user explicitly started the session with `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1`; otherwise use ordinary subagents, and don't propose teams unprompted.

## When Agent Teams Fit

With teams enabled, they fit when:

- The user asks for comprehensive code review (suggest 3-reviewer team)
- The task involves work across multiple layers (frontend + backend + database)
- The user describes a bug with unclear root cause (suggest competing hypothesis investigation)
- The user wants to evaluate multiple technologies or approaches (suggest research panel)
- The task can be clearly split into independent, file-separated workstreams

Do NOT suggest agent teams when:

- The task is sequential or involves same-file edits
- A single session or subagent would suffice
- The coordination overhead would exceed the benefit
- The user is on a tight token budget

## Team Templates

### Full Review (3 teammates)

Security reviewer + Performance reviewer + Correctness reviewer (edge cases, error paths, callers).
Each reviews independently, then lead synthesizes findings.

### Feature Dev (3 teammates)

Architect (plan approval required) + Implementer + Reviewer.
Architect plans first, implementer follows, reviewer validates the diff by review.

### Debug Squad (3-5 teammates)

Each teammate investigates a different hypothesis.
Adversarial debate structure -- teammates challenge each other's theories.

### Cross-Platform (3 teammates)

iOS developer + Android developer + Shared architect.
Each platform teammate owns their own directory. Architect coordinates API contracts.

### Full-Stack (3 teammates)

Frontend + Backend + Database.
Clear file ownership per layer. Coordinate on API contracts first.

### Research Panel (3 teammates)

Simplicity advocate + Performance advocate + Devil's advocate.
Each evaluates from a different perspective, then synthesize.

## Spawn Prompt Formula

Every spawn prompt should include:

1. Role: "You are the [role] teammate responsible for [domain]"
2. Files: "You own files in [directories]. Do not modify files outside your scope."
3. Focus: "Specifically look for / implement [details]"
4. Context: "[Project-specific information the teammate needs]"
5. Deliverable: "Report your findings as [format] / Implement [specific output]"

## Cross-Platform Awareness

- macOS and Linux: full support (in-process + tmux split panes)
- Windows: in-process mode only (automatic with auto setting)
- Always use `--teammate-mode auto` (default) for automatic detection

## Related Documentation

- `~/.claude/docs/AGENT-TEAMS.md` - Full guide with examples
- `~/.claude/docs/reference/workflows/agent-teams.md` - Decision framework
- `~/.claude/commands/assemble-team.md` - Quick team assembly command

© travisjneuman · MIT · travisjneuman/.claude

Similar skills

All skills for Agent setup: memory, context, orchestration →

agent-teams by other authors comes first.

agent-teams alinaqi agent-teams coordinates AI agents around a required TDD pipeline. Use it when feature work needs ordered checks, security review, and pull request creation. CX Codex +1 300installs agent-teams Rohit Ghumare agent-teams is a Claude Code skill that coordinates multiple sessions through shared tasks, messaging, and file locks. Teammates work in parallel on separate parts of the codebase. CX Codex +1 ★ 2,966repo stars ae-team ThinkingAIAgenticEngine ae-team — a reference skill for ae-cli: manage agent teams and run TeamRuns. Useful when the agent should find a team itself and carry a task through to artifacts. CX Codex +1 995installs team Yeachan-Heo The team skill for Claude Code splits a larger task among agents and tracks their work on a shared task list. Use it to coordinate parallel changes across a project. CX Codex +1 964installs multi-reviewer-patterns Seth Hobson The multi-reviewer-patterns skill coordinates parallel reviews across code-quality dimensions and merges the results into one report. Use it when a change needs several review angles. CX Codex +1 9,800installs team-spawn Seth Hobson team-spawn — a slash command for Claude Code, Codex, and Cursor that assembles a multi-agent team from a preset or custom composition. Seven presets cover review, debug, feature work, full-stack, research, security audit, and migration. CX Codex +1 ★ 42,780repo stars suede-agent-teams JasonColapietro suede-agent-teams splits development tasks across multiple AI agents with Git collision detection, exclusive file lanes, and structured ledger reporting. CX Codex +1 ★ 121repo stars claude-delegate amElnagdy The claude-delegate skill for Claude Code hands a coding task to a separate Claude session, then guides review of its changes. It activates only when delegation is explicitly requested. CX Codex +1 6,700installs

From the same set117

The whole set →
generic-fullstack-code-reviewer generic-fullstack-code-reviewer is a Claude Code skill for reviewing full-stack changes before commits and during pull request reviews. It checks Next.js and NestJS code for bugs, security issues, and project-rule mismatches. CX Codex +1 157installs seo-analytics-auditor seo-analytics-auditor — a skill that checks meta tags, structured data, analytics scripts and Core Web Vitals on a page. It reads the source HTML and walks through the SEO checklist item by item. CX Codex +1 156installs financial-scenario-planner financial-scenario-planner runs a financial plan through bear, base and bull cases, then adds sensitivity tables, break-even and a runway check. Useful when the assumptions behind a model need testing. CX Codex +1 156installs product-management product-management equips your agent with product strategy frameworks: RICE scoring, roadmap horizons, OKR setups, and PMF validation surveys. CX Codex +1 153installs generic-fullstack-ux-designer generic-fullstack-ux-designer is a Claude Code skill that guides UX design for Next.js and NestJS apps, covering loading states, optimistic UI, forms, and SSR quirks. CX Codex +1 153installs fundraising-analyzer The fundraising-analyzer skill breaks down nonprofit fundraising with donor segments, campaign ROI, retention metrics, and trend views. Use it to review results or plan campaigns. CX Codex +1 153installs microservices-architecture The microservices-architecture Claude Code skill guides microservice design, monolith decomposition, and service communication choices. It activates for distributed-system design tasks. CX Codex +1 153installs debug-systematic debug-systematic — a four-phase debugging skill for hard-to-reproduce failures, race conditions, and unexplained behavior. It guides the agent from reproduction to verified fix. CX Codex +1 152installs

Project sponsors

Companies that keep this analytics open