What
An MCP server is an add-on tool server that Claude Code connects to. Elicitation is a server asking you for input through Claude Code. When Claude Code connects, it tells each server what it supports. MCP server configs now accept a new optional true/false setting, bareElicitationCapability, in every connection type: stdio, sse, http and ws.
When it is set to true, the server gets the plain capability list instead of the legacy URL-elicitation version, so elicitation support is not announced to that server. The setting is only read when the tengu_mcp_legacy_url_elicitation flag is on. That flag's built-in fallback is off, so without a server-provided value the setting has no effect. For that flag, the flag server returned off for this site's account and for the anonymous baseline, but no reading has been taken under this release yet. Nothing has been read about the related tengu_mcp_url_elicitation flag.
Why
This lets you turn off the elicitation announcement for a single MCP server that misbehaves when it sees it. In practice it only changes anything where the legacy flag is on.
tengu_mcp_legacy_url_elicitation Off in both readingsThe flag server returned off for the account this site reads and for the anonymous baseline. A reading of off cannot rule out a rollout these two readings sit outside of.
This account: off · anonymous baseline: off · compiled default in v2.1.281: off
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.
tengu_mcp_url_elicitation Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.1.281: 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 v2.1.281. It isn't a statement about your account. What a flag value here can and cannot tell you
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
* Added URL prompts from MCP servers on the 2025-11-25 protocol, for example to sign in. If a server no longer connects after this update, add "bareElicitationCapability": true to its MCP config entrychangelog see the edit
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.
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.
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.
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.
Anthropic's documentation has since written up bareElicitationCapability, on Claude Code changelog.
New in this build: tengu_mcp_legacy_url_elicitation