Multi-Project Workflow

Clint is designed to manage Claude Code sessions across many projects simultaneously. Here’s how to use it effectively.

The HQ Pattern

The core pattern is:

1. Clint HQ runs permanently — it’s your command center
2. Project sessions are spawned on demand, each in their own tmux session
3. All sessions are reachable via claude.ai/code and optionally Telegram

clint-hq (always on)
    |
    +-- clint-my-api         (spawned for ~/work/my-api)
    +-- clint-my-frontend    (spawned for ~/work/my-frontend)
    +-- clint-my-api-fix-bug (spawned for worktree)

Day-to-Day Workflow

Morning: Start HQ

clint start

Or if it’s already running, verify:

clint status

Spawn Sessions as Needed

From terminal:

clint spawn my-api
clint spawn my-frontend --new-worktree feat-redesign

From Telegram:

“Open sessions for my-api and my-frontend”

From claude.ai/code:
Connect to the HQ session and ask Clint to spawn what you need.

Work Across Projects

Each session at claude.ai/code is independent. You can:

• Have my-api working on a backend feature
• Have my-frontend implementing the UI for it
• Ask the HQ session to coordinate between them

End of Day

# Stop specific sessions
clint stop my-frontend

# Or stop everything
clint stop-all

Project Groups

If you have a monorepo or org folder with multiple repos inside:

~/work/my-org/
├── service-a/
├── service-b/
└── shared-lib/

Configure it as a group:

[projects.my-org]
type = "group"

Now clint list shows:

  my-org/service-a    (git)
  my-org/service-b    (git)
  my-org/shared-lib   (git)

And you can spawn sessions for them:

clint spawn my-org/service-a

Session Naming

Sessions follow a predictable naming convention:

This makes it easy to find sessions in tmux, clint status, and claude.ai/code.

Capacity

By default, Remote Control allows 32 concurrent sessions. Adjust in config:

[hq]
capacity = 16

Or per-start:

clint start --capacity 64