[Discussion] Proactively detect and surface server-side rate limiting

Which client?

All / shared core

The problem

When a connected server starts rate-limiting requests, the Inspector currently just shows the raw error/timeout, without any recognition that a batch of failures might be rate-limit-shaped (e.g. repeated 429s or a consistent latency cliff), leaving distinguishing 'server down' from 'server throttling' up to the user.

Solution you have in mind (optional)

Recognize common rate-limit signals (429 status, Retry-After header) and surface a specific, actionable message with the retry timing instead of a generic error. Flagging as a discussion since it touches error-handling UX across all three clients.

Alternatives or workarounds you have tried

No response

Already built it locally?

No response

Before you submit

• I searched existing issues and this is not a duplicate.
• This is a request for the Inspector itself, not for the MCP specification or an SDK.

No files or tests are named. Start by tracing the shared error handling used by all three clients, then inspect how 429 responses and Retry-After headers are represented. Done means rate-limit-shaped failures produce a specific actionable message with retry timing rather than only a generic error or timeout.