Anthropic has rolled out a new feature that lets separate Claude Code sessions communicate directly with each other, eliminating the need for developers to manually relay information between parallel tasks. The feature, announced on August 7, 2026, uses plain-text messaging over private local channels to keep sessions coordinated without sharing conversation history, files, or sensitive context.

What Problem Does This Actually Solve?

Anyone running multiple Claude Code sessions knows the friction well. Imagine one session migrating a database while another builds an API that depends on it, and a third writes tests against both. When the migration session renames a column, the API session has no idea. It keeps building against a name that no longer exists, and you don't find out until something breaks twenty minutes later.

Until now, you were the message bus. You read the first session's output, understood it, switched terminals, retyped the important part, and hit enter. It was slow, lossy, and easy to forget when you were deep in something else. This was such a common pain point that it became an open feature request on Claude Code's issue tracker, with developers describing the problem in monorepos and microservice architectures where an orchestrator and worker workflows needed to coordinate.

How Does the Messaging Actually Work?

The mechanics are deliberately simple. When one session learns something another session needs, Claude writes a short note and delivers it. That's it. A message is plain text, nothing else. Not your conversation history, not your files, not context. Just words that one Claude wrote for another Claude to read.

The documented example is a session that finished updating the database setup, telling another session what new piece of data it added and that it is now safe to catch up. Short. Factual. Actionable. Claude can decide to send one on its own when it notices the need, or you can ask in plain English: tell the session working on the payments API what we just changed. You don't write the message. Claude does. You give it the intent and it phrases the note.

What Are the Technical Requirements and Limitations?

The feature requires Claude Code v2.1.224 or later and is currently available only on macOS and Linux, including Linux running inside Windows Subsystem for Linux 2 (WSL 2) on Windows machines. Native Windows does not support it yet. The feature is also not available if you run Claude Code through Amazon Bedrock, Claude Platform on AWS, Google Cloud's Agent Platform, or Microsoft Foundry. It is an Anthropic API feature only.

Local sessions in the Claude Desktop app's Code tab can use the feature on supported operating systems, and when they do, the desktop app can see and message terminal sessions, which its own built-in session tools cannot do. However, there is a silent kill switch: if you have set any of four environment variables to turn off feature-flag checks, cross-session messaging stays off quietly. Many privacy-conscious developers have one of these set and will spend an afternoon wondering why nothing works.

How to Set Up and Use Cross-Session Messaging

• Name Your Sessions: Sessions are named either with the /rename command or the --name flag when launching. If you don't set a name, Claude Code names it after the folder you are working in, something like myapp-3f. To see who is reachable, run /list-agents or /peers.
• Understand Message Delivery States: Messages can be delivered immediately to the other Claude, held and set aside with a notice until you approve them or a setting changes, or refused and dropped entirely. Timing matters: if the receiving session is mid-task, Claude reads the message between tool calls without interrupting a running command. If the session is idle at the prompt, the message starts a new turn immediately.
• Check Your Version and Environment: Run claude --version to verify you have v2.1.224 or later. If /list-agents is not recognized as a command, your session does not have the feature. If it runs but a message did not arrive, check whether any privacy-focused environment variables are silently disabling the feature.

What Are the Safety Boundaries?

Anthropic's documentation is unusually honest about what the feature cannot do, and these boundaries are worth understanding properly. Plain text only: no structured data, no files, no attachments, no context transfer. If you need to move a whole conversation, you resume the session instead. Messaging is not a context-sharing mechanism and was never designed to be one.

A message is not consent. A message from another session can never approve a pending permission prompt on your behalf. This is the single most important safety line in the whole design. If session A has been denied permission to delete a directory, it cannot ask session B to do it instead. Claude is explicitly instructed to route that work back to you. The feature also cannot change your configuration, edit permission settings, rewrite CLAUDE.md, or change anything about how the receiving session behaves. Commands do not run: if a message contains something like /compact, it arrives as plain text and Claude Code will not execute it. Permission prompts still fire if acting on a message needs approval.

What Has the Early Reception Been?

Multiple distinct users immediately flagged the native Windows gap when the feature launched. The cross-session teamwork arrives in v2.1.224 but stays macOS and Linux only, and this limitation appeared prominently in discussions across Windows forums and international threads under the announcement. Enterprise provider restrictions appear mainly in the documentation rather than in widespread public complaints so far, since the feature was only hours old at the time of reporting. The four environment variables that silently disable the feature are documented, though early public reports of people tripping over them are sparse.

The real value proposition is straightforward: the receiving session picks the message up mid-task, so the correction lands before you notice the problem. For developers managing complex workflows across multiple Claude Code sessions, this eliminates the manual coordination tax that has plagued parallel development work since the tool's inception.