Documentation
Everything you need to run a shared Claude Code session with your team: install the CLI, start a session, bring the mob in, and scale it to your organization.
Getting started
- What is MobSession?
MobSession is mob programming for AI coding agents: one orchestrator runs a shared Claude Code agent and a mob of teammates drives it together, live.
- Quickstart: your first session
Install the MobSession CLI, start a shared Claude Code session, and bring a teammate in to send the first prompt.
- Joining a session
Open a share link, pick a name, and steer a live MobSession from your browser. No install, no account.
- The session interface
A tour of the MobSession web UI: header, presence, transcript, tool cards, turn dividers, composer, reactions, and the orchestrator's terminal view.
Core concepts
- The orchestrator, the mob, and the agent
The core mental model behind MobSession: three roles, who controls what, and why the agent runs on your real machine.
- How prompts queue and run
MobSession's one-turn-at-a-time model: prompts queue in submission order, there is exactly one source of truth, and how to interrupt the current turn.
- Attribution: who steered what
Every prompt and reaction in a MobSession is tagged with the sender's display name and a stable per-person color, so the team can see who is steering the agent.
- Terminal and web, in sync
How a MobSession keeps the orchestrator's terminal and the mob's browser as one live, bidirectional session.
Working as a team
- Mob programming with MobSession
Mob programming puts the whole team on one problem at one keyboard. MobSession keeps that shape and makes an AI coding agent the driver while your team navigates.
- Pair programming, amplified
MobSession lets two engineers steer the same AI agent from a shared link, with live driver and navigator roles and no screen-share.
- Onboard engineers in live sessions
Ramp new hires by steering the agent through your real codebase together, instead of handing them stale docs.
- Mob an incident
Put multiple responders on one live investigation, steering a single shared agent that reads logs, traces the bug, and proposes a fix while everyone watches.
- Review at write-time
Steer the agent live as code is written, attribute every contribution, and keep pull-request review as the final gate.
For enterprises
- MobSession for enterprises
Why MobSession fits enterprise engineering teams today, plus a map of the enterprise capabilities rolling out in early access.
- Accounts, organizations, and teamspreview
How MobSession groups people into organizations and teams, scopes sessions to a team, and builds on today's link-based joining.
- SSO and access controlpreview
Sign in with corporate identity, replace self-entered names with verified identity, and gate sessions with role-based access control.
- Audit logs and compliancepreview
Attribution today, the durable audit trail rolling out for teams, and where our compliance program stands.
- Session history and mob ROIpreview
Every MobSession is saved as a durable record: transcript, contributors, files, pull requests, time, and cost, plus an ROI report you can view in /account and export.
- Data handling and retentionpreview
Exactly what stays on your machine, what is sent to the session backbone, and the retention controls rolling out for teams.
- Self-hosting MobSessionpreview
Run your own MobSession backend and web app inside your network so session data stays in your environment.
Reference
- CLI reference
Install the MobSession CLI, start a session, and drive it from your terminal: commands, keys, and requirements.
- Security model
How access, execution, and control work in MobSession today, and the controls rolling out for teams.
- Frequently asked questions
Honest answers to what MobSession is, where the agent runs, who can join, and how your code and data are handled today.