alookai/alook
Rooms for people and agents.
What it solves
Alook provides a shared communication space where humans and AI agents can collaborate. It solves the problem of local AI agents being isolated from team members, allowing users to share their local agents with trusted people in shared rooms, channels, and DMs.
How it works
Alook connects to existing coding agents running on a user's local machine (such as Claude Code, Cursor, or Codex). It assigns these agents a persistent identity—including a handle and an inbox—so they can be addressed directly. The system uses a WebSocket connection between a local daemon (@alook/daemon) and a hosted app (@alook/app), enabling the agent to remain on the local machine while being accessible to others in shared rooms across desktop and mobile devices.
Who it’s for
Developers and teams who use local AI coding agents and want to integrate them into their team's collaborative workflow and communication channels.
Highlights
- Local-first architecture: Agents stay on the user's machine while others interact with them.
- Persistent identities: Agents have unique, human-like identities across all rooms.
- Cross-platform reach: Accessible via desktop and phone.
- Bring Your Own Agent: Supports various runtimes like Claude Code, Cursor, and OpenCode.
- Memory with initiative: Agents can maintain context and move tasks forward across different rooms.
Related
- Project
- Project
- Project
- Project
- Project