The switch that lets stdio MCP servers (local programs Claude Code talks to over standard input and output) use protocol-version negotiation now falls back to on when the server sends nothing. Before, it fell back to off. The flag server also returned it on for this site's account and for an account with no special settings, first seen on 2026-09-29.
Claude Code can also remember what it learned from probing servers with server/discover, which is how it asks a server what it supports. It reads this from a file and ignores the file if:
- it is not a regular file
- it is larger than a size limit
- it cannot be read
When the file is ignored, Claude Code notes this in its log and probes servers as if nothing had been remembered. Entries for stdio servers are kept apart from the others.
Local MCP servers can now get protocol negotiation without anyone having to turn it on remotely, and a damaged or oversized memory file cannot get in the way of connecting.
tengu_mcp_protocol_negotiation_stdio On for this account, and not off by defaultThe flag server returned on for the one account this site reads, and nothing in this release compiles it off by default. The compiled default is shown below, and says which it is when we cannot read one: a fifth of gates compile in a string or a number rather than on or off, and most published releases have no gate table behind them at all. No client can see what the server returns for your account.
This account: on · anonymous baseline: on · compiled default in v2.1.292: on
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.1.292. It isn't a statement about your account. What a flag value here can and cannot tell you
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.
A reading is one sample. Claude Code evaluates its flags remotely, so no client sees the targeting rule behind a value and this says nothing about your account.