View on GitHub

AgentTeamsAI

Get a lot done by doing very little

You're the Boss, agents are your team. They handle tasks themselves, message each other, review each other's code. You just look at the kanban board and drink coffee.

Supported AI providers

75+
Providers
200+
Models
  • Claude Code
  • Codex
  • OpenCode
  • Cursor
  • SuperGrok
  • GitHub Copilot
  • Z.AI
  • MiniMax
  • Kiro
Download NowFor your platformDocumentationGuides and setup
Watch Demo
Watch Demo
01
Give the Team a Goal
Agents break it into tasks and start moving without babysitting.
02
Kanban That Updates Itself
Cards shift as agents build, test, review, and unblock each other.
03
Bring Your AI Stack
Mix Claude Code, Codex, OpenCode, Cursor, SuperGrok, Copilot, Z.AI, MiniMax, and Kiro teammates.
04
Stay in the Loop
Jump in with comments, approvals, direct messages, or quick actions.
05
Your Machine, Your Code
Run commands in the built-in terminal and track logs, file changes, and every agent's work.

See it in action

Real screenshots from the app — kanban board, code review, agent teams, and more.

Kanban board with team messages and agent tasks
Token usage, costs, runs, and budget analytics
Select AI providers and models for each teammate
Connected AI coding providers and subscription limits
Organization hierarchy with teams, agents, and active tasks
Task discussion and review comments
Team members with roles, runtimes, and models
MCP server catalog and diagnostics
Responsive team notification settings and built-in triggers
Code review with file-level and hunk-level controls
Task details, attachments, and execution logs
Agent execution log with tool calls
Organization structure map with team and task details

"Your agents. Your teams. One AI organization."

Mix Claude, Codex, Gemini, and OpenCode-connected providers in one team. Just a few examples: team hierarchies, Kanban, built-in messaging, a visual terminal, Solo Mode, budgets, flexible notifications, and automatic provider readiness checks.

Agent Teams

Build mixed teams with Claude, Codex, Gemini, and OpenCode-connected providers. Give agents distinct roles, let them work autonomously in parallel, and coordinate across team boundaries.

Kanban Board

Agents manage the cards themselves: they pick up tasks, update statuses, and move work across the board. You do not have to touch the Kanban manually, but you stay in control.

Changes Viewer

Review task and teammate changes in one Cursor-style workspace. Navigate files and full diffs, accept or reject whole files or individual hunks, edit when needed, undo decisions, and send feedback.

Token Analytics & Budgets

See where tokens and estimated costs go across teams, agents, tasks, projects, models, and runs. Spot expensive work, track trends and forecasts, set monthly token or cost budgets, and get alerts at 80% and 100%.

Organizations & Global Overview

Organize teams into departments, squads, or any nested structure. See every organization on one live map with team and agent statuses, task progress, dependencies, delegation, and cross-team communication.

Terminal

Run commands in a visual terminal workspace with team and local shells, persistent history, autocomplete, and terminal settings.

Cross-Team Communication

Agents message each other within and across teams. Direct messaging, task comments, and quick actions.

Solo Mode

Start with a single agent that self-manages tasks. Expand to a full team whenever you need more power.

Download

Choose your platform and start building with AI agent teams.

macOS

Apple Silicon / Intel
Apple Silicon / Intel

Windows

64-bit / ARM64
64-bit / ARM64

Linux

64-bit
Download

How we compare

Feature-by-feature comparison with other AI coding tools.

FeatureAgent TeamsGas TownPaperclipCursorClaude Code CLI
Live team collaboration
✓█████████░ 9/10Live agents plan, delegate, work, and review together
✓████████░░ 8/10Persistent teams with coordination and recovery
◒███████░░░ 7/10Durable scheduled and event-triggered agents, less live peer teamwork
◒██████░░░░ 6/10Parallel agents + subagents, no peer team
◒███████░░░ 7/10Experimental teams + cross-session messaging; recovery limits
Human control and approvals
✓Per-action approvals, roles, and notifications
✓Gates, roles, escalation, and recovery
✓Board approvals, roles, pause, and stop
◒Command controls and admin policies
✓Permissions + hooks
Visual team progress
✓Teammates, tasks, blockers, handoffs, activity, logs
◒Agent tree + feed panels
◒Org chart/status, not a task/log map
◒Agents Window, no shared peer-team map
◒Terminal team panel, no graphical UI
All-in-one team workspace
✓Tasks, code, terminal, review, and teammates in one app
◒Mail/feed/dashboard across tools
◒Board + task chats, less live teammate view
◒Agents Window, no peer team workspace
◒Terminal agent view, no graphical workspace
Easy setup
✓Start free, no signup or API key
✗Manual CLI stack
◒npx onboarding + browser app
◒App install + account
◒CLI + env flag
Agent readiness and recovery
✓Clear ready/stuck status with launch diagnostics
✓Working, stalled, and failed health with recovery controls
◒Run status and orphan recovery
◒Agent status in Agents Window
◒Agent view statuses, logs, stop/respawn in terminal
Kanban board
✓5-column task board with review states and drag-and-drop ordering
✗Dashboard, not Kanban
✓7 columns, drag-and-drop
✗
✗
Code review
✓Built-in Changes viewer with file/hunk accept, reject, edit, undo, and feedback
◒Merge queue, no diff UI
◒Review gates, not inline code review
✓Local Agent Review + PR Bugbot
◒Agent review, no review UI
Team and agent communication
✓Direct agent messages and shared task links across teams
◒Mailboxes + handoffs
◒Comments + @mentions
N/A
◒Cross-machine/cloud session messaging, no shared cross-team task graph
Linked tasks
✓Tasks can link to and block each other
✓Task deps + grouped work
✓Goals, parent tasks, blockers
✗
✓Shared task list
Agent activity and history
✓Agent messages, tool calls, timeline, token use, and cost
◒Session recall, feed, metrics
◒Run transcripts + cost audit
◒Agent chat, diffs, artifacts, and cloud diagnostics
◒Transcripts, background logs, /usage, and /insights
Organizations & Global Overview
✓Nested organizations with live team, task, relation, and communication map
◒Roles + escalation, no org chart
✓Org chart + board governance
◒Team admin only
✗
Mixed AI teammates
✓Provider-aware teams with per-member models and mixed OpenCode side lanes
✓Many providers, terminal-first
✓Bring your own agents/runtimes
◒Multi-model parent + subagents, no peer team
◒Claude-only experimental teams
Budget controls
✓Monthly budget alerts + optional cap for supported scheduled one-shot runs
◒Cost tiers + digest, no hard caps
✓Per-agent budgets + hard stops
✓User/team spend caps + cloud-agent spend limits
✓Org/member spend limits + print-mode hard cap
Separate agent workspaces
✓Optional separate workspace per teammate
✓Core primitive
✓Worktrees / branches
✓Agents Window worktrees
✓Built-in for sessions and subagents
Terminal
✓Built-in visual terminal for team and local commands
◒Terminal-based workflow, no built-in terminal
◒Runs commands, no interactive terminal
✓Built-in IDE terminal
◒Runs in your terminal
Built-in code editor
✓With Git support
✗
✗Control plane, not editor
✓Full IDE
✗
Task files and context
✓Auto-attach, agents read & attach
✗Not task-level
✓Docs, attachments, work products
◒Prompt context, not task attachments
◒Prompt files/images, not task attachments
Price
FreeOSS + free model with no auth, paid providers optional
FreeOSS, runtime plans needed
FreeOSS, self-hosted + infra
Free + paid usage
Claude plan or API usage

Live collaboration scores are qualitative editorial assessments of documented communication, task ownership and dependencies, completion, and review capabilities, not benchmark results.

Agent Teams product facts checked in local source on September 22, 2026; competitor sources checked on September 22, 2026: Agent Teams organizations feature, Agent Teams mixed-provider team lanes, Agent Teams cross-team messaging, Agent Teams Kanban board, Agent Teams terminal workspace, Agent Teams token usage budgets, Agent Teams budget alerts, Agent Teams scheduled budget cap, Agent Teams runtime/provider readiness detection, Agent Teams local provider autodetection, Agent Teams Solo Mode, Agent Teams Changes viewer, historical research notes (June 25, 2026), Gastown README, Gastown provider guide, Gastown scheduler, Gastown dashboard source, Gastown release, Paperclip README, Paperclip adapters, Paperclip heartbeat protocol, Paperclip org chart, Paperclip OrgChart source, Paperclip budgets, Paperclip runtime services, Paperclip Kanban source, Paperclip work products, Paperclip release, Cursor terminal, Cursor Cloud Agents, Cursor Agent Review, Cursor Bugbot, Cursor worktrees, Cursor Agents Window, Cursor subagents, Cursor Models & Pricing, Cursor Team Pricing, Cursor pricing, Claude Code CLI, Claude Code agent teams, Claude Code cross-session messaging, Claude Code worktrees, Claude Code subagents, Claude Code workflows, Claude Code costs, Claude pricing, Claude Code release.

Free to install. Free model included.

Start immediately with a free model with no auth - no account, API key, or credit card. Connect Claude Code, Codex, OpenCode/OpenRouter, Cursor, SuperGrok, GitHub Copilot, Z.AI, MiniMax, or Kiro only when you want more models.

Free

Free

$0

Start with the included free model with no auth. No signup, no API key, no credit card.

Free model with no auth included
  • Free model with no auth for first runs
  • No account or provider login required to try
  • Optional access to Claude Code, Codex, OpenCode, Cursor, SuperGrok, GitHub Copilot, Z.AI, MiniMax, and Kiro
  • Unlimited agent teams
  • Kanban board with real-time updates
  • Code review with diff view
  • Cross-team communication
  • Solo & team modes

Agent Teams has no paid tier. The free model with no auth lets you try it right away; paid provider usage is optional and controlled by the provider you choose.

Got questions? We've got answers

Everything you need to know about Agent Teams