Minor Changes
-
#2929
40f8f4eThanks @claude! -requireBearerAuthandverifyBearerTokentake a new optionalexpectedResource, which makes them accept only tokens issued for this resource (the token's audience). Set it to the value your authorization server puts into tokens meant for this server, usually the server's URL. When it is set, a token is accepted only if the verifier reports that value inAuthInfo.resource; the two are compared as strings, ignoring a fragment and one trailing slash. A token reported for another value, or for none, is answered401 invalid_tokenwith the usualWWW-Authenticatechallenge. When it is not set, nothing changes. To use it, passexpectedResourceand haveverifyAccessTokenfillAuthInfo.resource, for example from theaudclaim. The option is declared on a new exported type,VerifyBearerTokenOptions, which extendsBearerAuthOptions;BearerAuthOptionsitself is unchanged. The ExpressrequireBearerAuthpasses the option through. With Express,@modelcontextprotocol/expresshas to be upgraded to this release as well: 2.0.1 does not pass the option on, so nothing is compared. Its options type does not have the option, so TypeScript reports anexpectedResourcewritten in a call to the 2.0.1requireBearerAuthas an error. -
#2926
6d8dbc6Thanks @claude! -McpServernow accepts amaxToolInputElementsoption that limits the number of elements in tool-call arguments: the largest combined number of array elements and object members a singletools/callargumentspayload may contain. It is off by default, so behavior is unchanged unless you set it. When it is set and a call exceeds it, that call is answered with anisError: truetool result that names the limit, before the input schema runs, and the server keeps serving. Set it above the largest arguments your tools legitimately accept;maxRequestBodySizeremains the primary limit on request size. The value must be a number of at least 1, orInfinityfor no limit; any other value is rejected at construction. The options type is exported asMcpServerOptions. -
#2918
84804c2Thanks @claude! - AServerorMcpServernow serves one connection at a time, and a Streamable HTTP server transport without sessions (sessionIdGenerator: undefined) serves one request. An app that uses one server object, or one stateless transport, for every HTTP request fails on the second request after this upgrade. Build the server and the transport per request instead.What keeps working without a change:
createMcpHandler(buildServer)andserveStdio(buildServer), wherebuildServerreturns a new server on every call.- A handler that builds a new server and a new stateless transport for each request.
- One server and one transport per session (a transport with a
sessionIdGenerator). - Connecting a server again after
close(). Client.
What fails now, how it shows, and what to change:
- One server object with a new stateless transport per request (
const server = new McpServer(...)outside the handler,await server.connect(transport)inside it): the second HTTP request the process receives fails, and so does every later one.connect()rejects with anSdkErrorof codeALREADY_CONNECTED. If the handler closes the transport when the response ends, requests that arrive one after the other still work and a request that overlaps another one fails. Change: movenew McpServer(...)and its registrations into the handler. - One stateless transport for every request (a transport built once with
sessionIdGenerator: undefined): the second HTTP request fails.WebStandardStreamableHTTPServerTransport.handleRequest()rejects withStateless transport cannot be reused across requests. Create a new transport per request., andNodeStreamableHTTPServerTransport.handleRequest()answers500. Change: build the server and the transport inside the handler and connect them there. createMcpHandler(() => server)with a server built once: a request that arrives after the previous response has been read to its end still works. A request that arrives while another one is being served is answered500with the JSON-RPC error-32603(Internal server error); the reason is reported only through theonerroroption. Change: pass a function that builds the server, as increateMcpHandler(buildServer).- One server object for every session: the
initializerequest of the second session fails withALREADY_CONNECTED. Change: build a server per session.
What the caller sees when
connect()orhandleRequest()rejects depends on the host. Express 5, Fastify and Hono answer500. A plainnode:httplistener without its own error handling gets an unhandled rejection, which ends the process.The README examples of
@modelcontextprotocol/express,@modelcontextprotocol/fastify,@modelcontextprotocol/honoand@modelcontextprotocol/node, and the handler examples in the JSDoc ofWebStandardStreamableHTTPServerTransportandNodeStreamableHTTPServerTransport, now build a server and a transport per request. -
#2907
e55f9acThanks @claude! -allowedOriginsandvalidateOriginHeaderaccept lowercase entries of the form<scheme>://*, such asmoz-extension://*orchrome-extension://*, which admit every origin of that scheme. This lets a server admit MCP clients that run as a browser extension when the extension ID cannot be listed, as on Firefox, where it differs on every install.http://*andhttps://*are not honoured, and the defaults are unchanged.
Patch Changes
-
#2599
5238fbaThanks @freya0926! - A server can now serve, and a client can now call,tasks/getandtasks/cancelof the Tasks extension (SEP-2663) on a 2026-07-28 connection, when the handler is registered and the request is sent with an explicit schema. Every other method that a protocol revision removed is still refused. If one server factory serves both eras and such a handler is meant for 2025-era clients only, register it only whenctx.era === 'legacy'. -
#2107
2fc49eaThanks @pragnyanramtha! -prompts/getwithoutargumentsno longer fails with "Invalid arguments" when every argument of the prompt is optional. A missingargumentsis now validated as{}, as it already is fortools/call, so a top-level.optional()or.default(...)onargsSchemano longer seesundefined. -
#2889
4d94e7bThanks @claude! -registerToolno longer converts tool schemas up front, so a server built per request stops converting every tool on every request. The warning about an invalidx-mcp-headerdeclaration now appears each time tools are listed, not when the tool is registered. -
#2908
633dd3eThanks @claude! - Thelicensefield of the package manifests is nowApache-2.0; theLICENSEfile shipped in each package carries the full terms, including the MIT text for earlier contributions. No code change. -
#2841
2237555Thanks @sharziki! -McpServer.registerPrompt()now types the callback correctly when noargsSchemais given: its one parameter is the server context. Before, readingctx.mcpReqthere was a type error although it worked at runtime. Prompts registered with anargsSchemaare unchanged. -
Updated dependencies [
633dd3e]:- @modelcontextprotocol/core@2.3.0