Description

I'm working on an MCP tool which takes a while to boot.

When interacting with this tool through the standard I/O transport, this means that each new request takes a while to complete. That makes the interactive shell mode slower than it needs to be.

This change introduces an option for not closing the process after each request. (The SSE transport will already keep the connection alive.)

Type of change

• New feature (non-breaking change which adds functionality)

How Has This Been Tested?

Outside running the tests, I have used this locally with mcptools shell <command>.

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in downstream modules