How do I install the Agentteams skill?

Run npx skillmds@latest add tomevault-io/agentteams in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agentteams safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agentteams?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agentteams free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agentteams?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agentteams by tomevault-io — Agent Skill

Name: Agentteams
Author: tomevault-io

Agentteams

Mode B parallelism. Manually invoke Claude Code Agent Teams (multiple full Claude Code instances coordinating via shared task list + worktree isolation + Agent View). The multispec engine auto-selects this mode when the work demands it; this skill is for forcing it. Use when this capability is needed.

tomevault-io Updated Sep 17, 2026

File contents

/agentteams — Mode B parallelism (Claude Code Agent Teams)

Multiple full Claude Code instances coordinating via a shared task list with worktree isolation. Different from Mode A (SDK subagents in one query()) — these are separate processes, separate sessions, separate worktrees.

Requirements

Claude Code v2.1.32+ (/agent-teams shipped experimental).
CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=true (set automatically by .claude/settings.json).
Agent View available (Claude Code v2.1.139+) for live monitoring.

How auto-selection works

The multispec engine picks Mode B when ANY of:

5 sub-Specs
Estimated total time > 30 minutes
Cross-sub-Spec peer coordination required
Overlapping write sets (worktree isolation solves this)

Override with cmax run "..." --mode teams or cmax run "..." --mode solo.

Mechanics

Each sub-Spec spawns a background Claude Code session with the SPEC + shared task list seed.
Shared task list: .claudemax/state/agent-teams-<ts>/shared-task-list.md. Every teammate has read/write.
Worktree isolation: each session writes to .claude/worktrees/<session-id>/ automatically on first write.
Agent View dashboard: cd && claude --agent-view shows live state of all running sessions.

When to manually invoke

A multi-day swarm-style project where you want the coordination primitive even for < 5 sub-Specs.
Stress-testing the agent-teams flow on a small problem before trusting it on a big one.

Anthropic's proof point

16 agents wrote a Rust-based C compiler from scratch (100K LOC, ~2000 sessions, $20K credit) using this primitive. See https://www.anthropic.com/engineering/building-c-compiler.

Source: waitdeadai/claudemax — distributed by TomeVault.

---name: agentteamsdescription: Mode B parallelism. Manually invoke Claude Code Agent Teams (multiple full Claude Code instances coordinating via shared task list + worktree isolation + Agent View). The multispec engine auto-selects this mode when the work demands it; this skill is for forcing it. Use when this capability is needed.---# /agentteams — Mode B parallelism (Claude Code Agent Teams)Multiple full Claude Code *instances* coordinating via a shared task list with worktree isolation. Different from Mode A (SDK subagents in one query()) — these are separate processes, separate sessions, separate worktrees.## Requirements- Claude Code v2.1.32+ (`/agent-teams` shipped experimental).- `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=true` (set automatically by `.claude/settings.json`).- Agent View available (Claude Code v2.1.139+) for live monitoring.## How auto-selection worksThe multispec engine picks Mode B when ANY of:- > 5 sub-Specs- Estimated total time > 30 minutes- Cross-sub-Spec peer coordination required- Overlapping write sets (worktree isolation solves this)Override with `cmax run "..." --mode teams` or `cmax run "..." --mode solo`.## Mechanics- Each sub-Spec spawns a background Claude Code session with the SPEC + shared task list seed.- Shared task list: `.claudemax/state/agent-teams-<ts>/shared-task-list.md`. Every teammate has read/write.- Worktree isolation: each session writes to `.claude/worktrees/<session-id>/` automatically on first write.- Agent View dashboard: `cd && claude --agent-view` shows live state of all running sessions.## When to manually invoke- A multi-day swarm-style project where you want the coordination primitive even for < 5 sub-Specs.- Stress-testing the agent-teams flow on a small problem before trusting it on a big one.## Anthropic's proof point16 agents wrote a Rust-based C compiler from scratch (100K LOC, ~2000 sessions, $20K credit) using this primitive. See `https://www.anthropic.com/engineering/building-c-compiler`.---> Source: [waitdeadai/claudemax](https://github.com/waitdeadai/claudemax) — distributed by [TomeVault](https://tomevault.io).

Frequently asked questions