DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to run several coding agents in parallel worktrees with Worktrunk

Create a worktree per task and launch an agent in each with one command.

Short answer: Run brew install worktrunk && wt config shell install (or cargo install worktrunk && wt config shell install); then run wt switch -x claude -c feature-a -- 'Add user authentication'; run wt switch -x claude -c feature-b -- 'Fix the pagination bug'. All steps.

Who it is forA developer managing 5 to 10 agents who is tired of typing branch names three times.

The problemAgents in one directory step on each other's changes, and plain git worktree commands are clunky.

The outcomeOne worktree per agent, each started with a task, addressed by branch name.

What you need first

  • git
  • Homebrew or Cargo to install
  • Shell integration via `wt config shell install` so commands can change directories
  • An agent such as Claude Code or Codex

Steps

  1. Run brew install worktrunk && wt config shell install (or cargo install worktrunk && wt config shell install)
  2. Run wt switch -x claude -c feature-a -- 'Add user authentication'
  3. Run wt switch -x claude -c feature-b -- 'Fix the pagination bug'
  4. Run wt list to see the status of each worktree

Prompt for your agent

Install worktrunk, then for each of <tasks> run `wt switch -x claude -c <branch> -- '<task>'` so every agent gets its own worktree. Use `wt list` to report status.

Caveat. On Windows, winget installs it as git-wt to avoid a clash with Windows Terminal's wt command.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for max-sixty/worktrunk and "Run several coding agents in parallel worktrees"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer