Parallel Coding Agents with Git Worktrees and Playwright MCP

Melvin Vivas · X post · 2026-05-20 · Open on X

Topics: AI Dev Tools & Productivity, AI Agents, Tool Use & MCP · Level: intermediate

Summary

The creator now does all his work in git worktrees, so several coding-agent tasks can run in parallel in separate, clean checkouts. A script sets up each new worktree for startup and testing. Playwright MCP is installed so the coding agent can test UI changes in a browser by itself.

Key points

• Git worktrees let you check out several branches at once in separate folders, which keeps parallel work isolated
• Run one coding agent per worktree so tasks proceed in parallel without conflicts
• Automate worktree setup with a script that installs dependencies, sets up the environment, and gets the app ready to start and test
• Install Playwright MCP so the agent can drive a browser and check UI changes on its own
• Together this gives a cleaner workflow than switching branches in one checkout

Resources mentioned

• Git worktree · docs · git-scm.com · free
  Git command for keeping several working trees (checked-out branches) from one repository at the same time.
• Playwright MCP · repo · github.com · free
  Microsoft's MCP server that lets AI agents control a browser through Playwright for UI testing and automation.
  Also in: Building and Deploying a Full App with the Codex App, MCP Servers and Skills (Melvin Vivas on X · notes)

Try this

• Use git worktrees to run several coding-agent tasks in parallel
• Write a script that prepares a fresh worktree for startup and testing
• Install Playwright MCP so your coding agent can check UI changes by itself
• Build a worktree bootstrap script that creates a worktree, installs dependencies, copies env files, and starts the app for agent testing