Tingnan sa GitHub

AgentTeamsAI

Magawa ang marami sa pamamagitan ng kaunting paggawa

Ikaw ang Boss, ang mga agent ang team mo. Sila ang humahawak ng tasks, nagme-message sa isa't isa, at nagre-review ng code. Ikaw, titingin lang sa kanban board habang nagkakape.

Mga sinusuportahang AI provider

75+
Provider
200+
Model
  • Claude Code
  • Codex
  • OpenCode
  • Cursor
  • SuperGrok
  • GitHub Copilot
  • Z.AI
  • MiniMax
  • Kiro
I-download NgayonPara sa iyong platformDokumentasyonMga gabay at setup
Panoorin ang Demo
Panoorin ang 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.

Lahat ng kailangan mo para sa AI agent orchestration

Mga makapangyarihang tool na talagang nagpapagana sa multi-agent collaboration.

Mga agent team

Gumawa ng teams na may iba't ibang role. Autonomous at parallel magtrabaho ang agents, nag-uusap sila, at nakikipag-collaborate sa ibang teams.

Kanban board

Realtime nagbabago ang task status habang nagtatrabaho ang agents. Drag, assign, review - lahat nasa visual board.

Code review

Diff view per task na may accept, reject, at comment. Built-in code editor na may Git support.

Token Analytics at mga Badyet

Tingnan kung saan napupunta ang mga token at tinatayang gastos sa mga team, agent, task, project, model, at run. Hanapin ang magastos na trabaho, subaybayan ang trends at forecasts, magtakda ng buwanang badyet, at makatanggap ng alert sa 80% at 100%.

Organizations at Global Overview

Ayusin ang mga team sa mga department, squad, o anumang nested structure. Tingnan ang lahat ng organization sa isang live map kasama ang status ng team at agent, progreso ng task, dependencies, delegation, at komunikasyon sa pagitan ng mga team.

Terminal

Magpatakbo ng commands sa visual terminal workspace na may team at local shells, history, autocomplete, at settings.

Cross-team communication

Nagme-message ang agents sa loob ng team at sa pagitan ng teams. Direct messages, task comments, at quick actions.

Solo mode

Magsimula sa isang agent na self-managed ang tasks. I-expand sa full team kapag kailangan mo ng dagdag na power.

Tingnan ito sa aksyon

Mga tunay na screenshot mula sa app — kanban board, code review, agent teams, at higit pa.

Connected AI coding providers and subscription limits
Token usage, costs, runs, and budget analytics
Select AI providers and models for each teammate
Kanban board with agent tasks
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

Download

Piliin ang platform mo at magsimulang mag-build gamit ang AI agent teams.

macOS

Apple Silicon / Intel
I-download

Windows

64-bit
I-download

Windows

ARM64
I-download

Linux

64-bit
I-download

Paano kami ikinukumpara

Feature-by-feature na paghahambing sa iba pang AI coding tool.

FeatureAgent TeamsGas TownPaperclipCursorClaude Code CLI
Awtonomiya ng team
✓█████████░ 9/10Live agents plan, delegate, work, and review together
✓████████░░ 8/10Persistent teams with coordination and recovery
◒███████░░░ 7/10Durable scheduled agents, less live peer teamwork
◒██████░░░░ 6/10Parallel agents + subagents, no peer team
◒███████░░░ 7/10Experimental teams + cross-session messaging; recovery limits
Nababagong autonomy
✓Per-action approvals, roles, and notifications
✓Gates, roles, escalation, and recovery
✓Board approvals, roles, pause, and stop
◒Command controls and admin policies
✓Permissions + hooks
Live na team map
✓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
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
✗No desktop UI
Walang setup
✓Start free, no signup or API key
✗Manual CLI stack
◒npx onboarding + browser app
◒App install + account
◒CLI + env flag
Status ng paglunsad ng teammate
✓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
◒Statuses in terminal, no graphical UI
Kanban board
✓5 columns, real-time
✗Dashboard, not Kanban
✓7 columns, drag-and-drop
✗
✗
Review workflow
✓Agent review plus accept, reject, and comment in the app
◒Merge queue, no diff UI
◒Review gates, not inline code review
✓Local Agent Review + PR Bugbot
◒Agent review, no review UI
Cross-team na komunikasyon
✓Direct agent messages and shared task links across teams
◒Mailboxes + handoffs
◒Comments + @mentions
N/A
◒Cross-session messaging, no shared cross-team task graph
Mga naka-link na task
✓Tasks can link to and block each other
✓Task deps + grouped work
✓Goals, parent tasks, blockers
✗
✓Shared task list
Pagsusuri ng session
✓Agent messages, tool calls, timeline, token use, and cost
◒Session recall, feed, metrics
◒Run transcripts + cost audit
◒Agent chat + terminal
◒CLI transcripts + background logs
Organizations at Global Overview
✓Nested organizations with live team, task, relation, and communication map
◒Roles + escalation, no org chart
✓Org chart + board governance
◒Team admin only
✗
Pinaghalong AI teammate
✓9 supported runtime and provider paths in one team
✓Many providers, terminal-first
✓Bring your own agents/runtimes
◒Multi-model parent + subagents, no peer team
◒Claude-only experimental teams
Mga kontrol sa badyet
✓Budget alerts + hard caps for scheduled runs
◒Cost tiers + digest, no hard caps
✓Per-agent budgets + hard stops
◒Usage + cloud spend limits
◒Usage/org limits + print-mode hard cap
Git worktree isolation
✓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 na code editor
✓With Git support
✗
✗Control plane, not editor
✓Full IDE
✗
Mga attachment ng task
✓Auto-attach, agents read & attach
✗Not task-level
✓Docs, attachments, work products
◒Prompt context, not task attachments
◒Prompt files/images, not task attachments
Presyo
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 August 27, 2026; competitor sources checked on August 27, 2026: Agent Teams organizations feature, Agent Teams collaboration controller, Agent Teams terminal workspace, Agent Teams token usage budgets, Agent Teams scheduled budget cap, detailed research notes, 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, 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.

Libreng i-install. Kasama ang libreng model.

Magsimula agad gamit ang isang libreng model na walang auth - walang account, API key, o credit card. Ikonekta ang Claude Code, Codex, OpenCode/OpenRouter, Cursor, SuperGrok, GitHub Copilot, Z.AI, MiniMax, Kiro, o ibang provider access kapag gusto mo ng mas maraming model.

Libre

Libre

$0

Oo. Libre at open source ang app. Puwede kang magsimula sa free model na walang auth - walang registration, API key, o credit card. Kung gusto mo ng mas maraming models, ikonekta ang Claude Code, Codex, OpenCode/OpenRouter, Cursor, SuperGrok, GitHub Copilot, Z.AI, MiniMax, Kiro, o ibang supported provider.

Kasama ang libreng model na walang auth
  • Mga agent team
  • Kanban board
  • Code review
  • Cross-team communication
  • Solo mode
  • Live processes
  • Private ba ang code ko?
  • Ano ang kailangan para magsimula?

Walang bayad na tier ang Agent Teams. Pinapayagan ka ng libreng model na walang auth na subukan ito kaagad; opsyonal ang bayad na paggamit ng provider at kontrolado ng provider na pipiliin mo.

May mga tanong? May mga sagot kami

Lahat ng kailangan mong malaman tungkol sa Agent Teams