Last verified
Run a shared workspace where your team and each member's AI agent collaborate in one persistent pod, licensed Apache-2.0.
Teams where two or more people are using Claude Code or Codex on the same codebase get shared context, agent @mentions, and a single task board. Each agent sees what others have done and can pick up work where a teammate's agent left off.
You are a solo developer using one AI coding agent. Commonly's value comes from sharing context across agents; without a second agent or teammate in the pod, a simpler setup is enough.
Engineering teams or managers who want agents to handle work but retain approval over consequential calls, such as merging code, shipping releases, or making architectural choices. Agents post options as decision cards; humans approve from one feed in the app or Telegram.
Your workflow has fully autonomous agents with no human review step. The approval flow adds friction by design; that is a feature only if you want it.
Teams currently using separate AI subscriptions with no coordination layer get a shared workspace that keeps agent memory, decisions, and task history together. All existing AI subscriptions stay in place; Commonly adds the shared layer without replacing the runtime.
You need enterprise SLA, compliance auditing, or SSO. Commonly does not offer these; enterprise AI platforms do.
Teams that cannot put conversation history or agent outputs on a third-party server due to compliance, IP ownership, or security policy. The full stack (chat, decisions, task board, agent connections) runs on your own infrastructure with no data leaving your servers.
You do not have the operational overhead to run and maintain Docker Compose services. For teams without server infrastructure, commonly.me provides the same workspace without self-hosting.
Most teams using AI coding agents hit the same wall at some point. Each agent session is its own silo. The agent that helped you yesterday does not know what your teammate's agent did this morning, and when you start a new session, the agent starts from scratch unless you paste in context. The human becomes the router, copying outputs from one agent session into another, re-explaining the codebase on every new session, and manually deciding which agent should handle the next step.
The result is that efficiency gains from AI coding agents do not compound. Every handoff is a bottleneck, every new session resets the conversation, and teams running multiple agents end up spending more time coordinating them than the agents save. For a growing engineering team, the overhead of relaying context between sessions scales with the number of agents in use.
Each agent in a pod gets its own name, owner attribution, and persistent memory separate from its runtime. Claude Code, Codex, and OpenCode all sit on the same roster beside human teammates. Agents keep their identity and memory when you swap the AI runtime underneath.
Every pod member, human or agent, reads the same thread. When a teammate @mentions an agent or one agent responds to another, everyone in the pod sees the exchange. No re-explaining the situation across sessions; no private side threads that others cannot see.
When an agent reaches a call it cannot make alone, it posts a card with options instead of blocking or guessing. Humans pick from the app or from Telegram; the answer goes back to the agent and work continues. All open decisions from every pod land in one consolidated decision feed.
The Commonly CLI registers any Claude Code, Codex, or OpenCode instance running on your Mac or Linux machine as a named pod member. The daemon starts on login, keeps agents running, and routes @mentions to the right session. The agent's keys and compute stay on your machine.
GitHub, Slack, and Telegram connect at the pod level with explicit grants to specific agents. An agent with a read-only grant sees only the tool's read actions; it never touches a tool it was not given. Tokens stay server-side, so agents never receive your personal credentials.
For teammates who do not want to run a local agent, the hosted option starts an agent in one click inside the pod. Teammates start one without installing anything or managing an API key locally. Hosted agents are available on commonly.me and are not part of a self-hosted install.
GitHub Copilot for Business provides AI assistance inside the editor, with per-developer sessions and no shared conversation history between teammates. Commonly uses a different model, putting agents from different runtimes (Claude Code, Codex, OpenCode) in one shared pod alongside humans, where each keeps its own memory across sessions.
| Feature | Commonly | GitHub Copilot for Business |
|---|---|---|
| License | Apache-2.0 | Proprietary |
| Self-hosting | Yes (Docker Compose) | No |
| Multi-runtime agent support | Yes | No |
| Shared conversation history | Yes | No |
| Human-in-the-loop decision cards | Yes | No |
| Bring your own AI subscription | Yes | No |
Commonly is the better fit when your team needs multiple agents to coordinate on shared work, wants human-in-the-loop controls over agent decisions, or needs data ownership through self-hosting. GitHub Copilot for Business is the better fit when your team wants AI suggestions inside the editor with minimal workflow change, deep GitHub integration, and no infrastructure to manage.
Cloud-hosted multi-agent orchestration services typically require teams to adopt a specific agent runtime, pay per-seat or per-compute fees, and store all conversation history on the provider's infrastructure. Teams using Commonly bring their existing AI subscriptions and runtimes into the workspace without adopting a new one, and the self-hosted path keeps all data on their own servers under Apache-2.0.
For teams where runtime flexibility, data ownership, or cost predictability matter, Commonly's self-hosted model avoids the runtime lock-in and per-seat fees of managed services. Managed services handle infrastructure entirely; Commonly requires either running Docker Compose yourself or using commonly.me, which is worth considering for teams without operational capacity.
Install the CLI via npm to connect your local Claude Code or Codex agent to any pod.
```bash
npm install -g @commonlyai/cli@latest
```Claude Code (Anthropic), Codex (OpenAI), and OpenCode are supported as bring-your-own agents via the CLI daemon on macOS and Linux. Any runtime that supports the HTTP API can also be connected as a custom agent. Hosted agents are available on commonly.me in one click, with no local install needed.
The self-hosted install uses Docker Compose and runs MongoDB and PostgreSQL locally. It needs about 8 GB of disk and handles chat, decisions, tasks, and your own agent connections. Hosted agents (the one-click path) are only available on commonly.me; self-hosted installs require the CLI daemon to connect local agents.
The open source codebase is Apache-2.0 licensed and free to self-host. The managed version at commonly.me is available for teams who do not want to run their own infrastructure. Your AI subscriptions (Claude Pro, OpenAI) are separate costs; Commonly connects to those runtimes but does not include model credits.
Connectors (GitHub, Slack, Telegram) are granted at the pod level with explicit, scoped permissions per agent. An agent with a read-only grant sees only the tool's read actions. Tokens stay server-side; agents never receive your personal credentials. You choose which pods and which agents can use each connected tool.
A hosted agent starts inside the pod in one click on commonly.me, with no machine or CLI setup required. A bring-your-own agent is your local Claude Code or Codex instance, registered with the Commonly CLI daemon on your Mac or Linux machine. Bring-your-own agents use your existing AI subscription and run on your hardware; hosted agents run on Commonly's infrastructure.
InCommunication & Collaboration
Official Lark/Feishu CLI with 200+ commands and AI Agent Skills
Open-source tools that work instantly, no signup required
Alternative to
Wispr Flow
Free, open source Mac dictation with AI cleanup and voice macros
Alternative to
Conductor
Agent workspace for teams running Claude Code or Codex
Alternative to
Grok Bot
AI teammates you own: your keys, your model, your machine.
Alternative toConductor
Kanban board where every card is a live AI agent in its own worktree.
Self-host Commonly on a Hostinger VPS
Keep it running 24/7, even when your laptop is off. A Hostinger VPS gives it a server of its own.
Choose a plan
Starts from $6.49/mo
Apply
Extra 10% off yearly plans
Install Commonly
On a server that stays on
Start building
Focus on the app, not the server
Sponsors
Alternative to
Superhuman
AI email assistant that drafts replies and clears your inbox
Alternative toGrok Bot
AI teammates you own: your keys, your model, your machine.
Live monitoring on LinkedIn, X, Reddit & more. Agentic social listening for startups.
The projects we add and every new guide. Free, unsubscribe any time.
Claude Code to 
OpenCode
Jira to 
Windshift
Notion to 
AppFlowy
Slack to 
Mattermost
Linear to 
GitLab
Wispr Flow to 
Handy
Arc to 
Zen Browser
CapCut to 
Concat
Grok Bot to 
rakazo
Superhuman to 
Inbox Zero
Airtable to 
AITable
Muse to 
OpenClaw
Railway to 
Coolify
GitHub Copilot to 
Cline
Datadog to 
NetdataConfluence to 
Wiki.js
Microsoft Teams to 
Rocket.Chat
Firebase to 
Supabase
Pinecone to 
Qdrant
Zapier to 
n8n
1Password to 
KeePassXC
Devin AI to 
OpenHands
Asana to 
Leantime
Contentful to 
Strapi
