CC - Multi-Agent Workflows

So far, everything we've done has been a single-agent workflow — one terminal session, one agent, doing one thing at a time. But Claude Code supports running multiple agents at the same time, which opens up some pretty powerful possibilities.

Running Multiple Terminal Sessions

The simplest version of this is just opening multiple terminal windows or tabs side by side. In each one, navigate to your project and run claude. Each session is completely independent — different conversation, different context — but they're all scoped to the same project directory.

So you could have one terminal planning out a new feature while another terminal is implementing tests. Both running at the same time.

Just keep in mind: two agents touching the same codebase simultaneously can get complicated. It's manageable with two sessions, but be careful going beyond that. People that go down this route often end up either using git worktrees or keeping multiple copies of the same repo, both enabling you to run multiple sessions at the same time.

Subagents

There's also a more automated approach: subagents. Claude Code has a built-in Agent tool that lets Claude spin up and orchestrate its own team of agents internally.

For example, you could ask Claude to run three agents in parallel — one reviewing the code for security issues, one for performance, and one for readability — and synthesize the results.

Here's how it works under the hood:

1. The main agent spawns one or more subagents and sends each one a prompt with instructions

2. Each subagent runs in its own isolated context window — it only knows what the main agent tells it in its prompt

3. The subagents go off, do their research, read files, work through the problem

4. Each one returns a result back to the main agent

5. The main agent gets just the final output — all the intermediate reasoning stays in the subagent and gets discarded

Open

Open

If subagents are running in parallel, the main agent waits for all of them to finish before moving on — so the total time is determined by the slowest one. Subagents are powerful because they let Claude do a lot of work in parallel without bloating the main context window.

Workflow Review

• Manual: open multiple terminal sessions and orchestrate them yourself — good for a couple of independent tasks

• Automated: ask Claude to use subagents and it handles the orchestration — good when you want Claude to coordinate everything toward a single goal

To trigger automated subagents, just ask naturally:

Claude will handle the rest.

Next Steps

Next up: Model Context Protocol. We'll look at how to extend Claude Code's toolbox by connecting it to external services.

CC - Multi-Agent Workflows