Cross Agent MCP — Communication skill for Claude Code

cross-agent MCP

  

**Let your Claude Code session and your Codex thread talk to each other — live, in VS Code, without either one losing its memory.**

cross-agent MCP is a relay MCP server for two coding agents that are already running side by side in the same VS Code window: an active **Claude Code** session and an active **Codex** thread. Either one can hand a message to the other through `send_to_codex` / `send_to_claude`, and the bridge finds each product's **currently active session** from the transcript it leaves on disk and **resumes it** — instead of spawning a disposable new agent — so both sides keep their full existing context.

That's the difference from just running a second CLI by hand: neither side has to re-explain the task, and neither one loses the conversation it was already having. A few things this is useful for:

• Get a second opinion without leaving your conversation. Ask Codex to review or double-check Claude's plan, or the other way around, and keep working while it thinks.
• Hand off a long task and keep going. send_to_* is asynchronous — it queues the message and returns immediately. Whenever the peer's answer is ready, it arrives back as a new message in your own session.
• Watch it happen, not just read a log. With the two panel shims from section 3 installed, both directions render in VS Code's real chat panel like any other message, instead of just appending a line to a transcript file.
• Nothing is lost to a timeout. Because nothing blocks, a peer turn that takes ten minutes is fine — the reply lands whenever it lands.

A real example

One user runs a single Claude session as a **master** that directs about a dozen role-based sub-sessions — a Claude or Codex session per ro