fix: cancel a Streamable HTTP call by closing its stream (#2140)

The Cancel button POSTed `notifications/cancelled` on every transport. On
a 2026-07-28 Streamable HTTP connection that is the wrong signal — the
spec makes closing the request's SSE response stream the cancellation
signal there and says the notification is neither required nor expected —
so a spec-compliant server acknowledged it 202 and dropped it while the
tool ran to completion.

The SDK already implements the fork, off `transport.hasPerRequestStream`.
Every Inspector connection is wrapped in `MessageTrackingTransport`, which
did not forward that flag, so the SDK read `undefined` and took the stdio
branch on every client — the CLI and TUI included, whose real
`StreamableHTTPClientTransport` advertises it correctly.

The web client needed three more links: its browser transport answers for
the upstream transport that lives on the backend, so it advertises the
flag for `streamable-http` and applies the SDK's `requestSignal` to its
`POST /api/mcp/send`; and `/api/mcp/send`, held open for the whole call,
forwards the browser's disconnect to the upstream `transport.send`.

The test fixture's `toWebRequest` built a Web Request with no `signal`, so
no fixture server could observe a disconnect at all — without that, a test
asserting the cancel reaches the server passes vacuously.

stdio and SSE keep `notifications/cancelled`: they multiplex every request
over one shared channel and have no per-request stream to close. So does
the legacy era, which predates the mechanism.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013hzwzS8UvBsr4yZm7o7sev
Signed-off-by: cliffhall <cliff@futurescale.com>

clients/web/src/test/core/mcp/messageTrackingTransport.test.ts

@@ -12,6 +12,8 @@ class FakeTransport implements Transport {
   sessionId?: string;
   // Optional on the SDK Transport interface; stdio omits it, HTTP defines it.
   setProtocolVersion?: (version: string) => void;
+  // Optional too: Streamable HTTP advertises it, stdio and SSE say nothing.
+  hasPerRequestStream?: boolean;
   async start(): Promise<void> {}
   async send(message: JSONRPCMessage): Promise<void> {
     this.sent.push(message);
@@ -272,4 +274,19 @@ describe("MessageTrackingTransport.setProtocolVersion", () => {
     expect(() => tracked.setProtocolVersion("2025-06-18")).not.toThrow();
     expect(tracked.protocolVersion).toBe("2025-06-18");
   });
+
+  // #2140: the SDK reads `hasPerRequestStream` off the transport it was handed
+  // — always this wrapper — to choose between aborting a request's own stream
+  // and POSTing `notifications/cancelled`. A wrapper that answers for the base
+  // transport instead of forwarding its answer breaks Cancel on every client.
+  it("forwards the base transport's per-request-stream capability", () => {
+    const { tracked, base } = makeTracked();
+    expect(tracked.hasPerRequestStream).toBeUndefined();
+
+    base.hasPerRequestStream = true;
+    expect(tracked.hasPerRequestStream).toBe(true);
+
+    base.hasPerRequestStream = false;
+    expect(tracked.hasPerRequestStream).toBe(false);
+  });
 });

clients/web/src/test/core/mcp/remote/remoteClientTransport.test.ts

@@ -569,4 +569,123 @@ describe("RemoteClientTransport", () => {
 
     await transport.close();
   });
+
+  // #2140: the Cancel button's wire signal is chosen by the SDK from the
+  // transport's `hasPerRequestStream`. The real upstream transport lives on the
+  // backend, so this transport answers for it — and must answer per server
+  // type, since only Streamable HTTP gives a request a stream of its own to
+  // close.
+  describe("per-request-stream cancellation (#2140)", () => {
+    it("advertises hasPerRequestStream only for streamable-http", () => {
+      const forType = (c: MCPServerConfig) =>
+        new RemoteClientTransport(
+          { baseUrl, fetchFn: vi.fn() as unknown as typeof fetch },
+          c,
+        ).hasPerRequestStream;
+
+      expect(
+        forType({ type: "streamable-http", url: "http://server.example/mcp" }),
+      ).toBe(true);
+      // stdio and SSE multiplex every request over one shared channel, so
+      // aborting anything would cancel the whole session rather than the call.
+      // The SDK must keep sending `notifications/cancelled` for them.
+      expect(forType(config)).toBe(false);
+      expect(forType({ type: "sse", url: "http://server.example/sse" })).toBe(
+        false,
+      );
+    });
+
+    it("applies the SDK's requestSignal to the browser-to-backend send fetch", async () => {
+      const seenInits: (RequestInit | undefined)[] = [];
+      let sse = createPushableSseStream();
+      const fetchFn = vi
+        .fn<typeof fetch>()
+        .mockImplementation(async (input, init) => {
+          const url = String(input);
+          if (url.endsWith("/api/mcp/connect")) {
+            return new Response(JSON.stringify({ sessionId: "abc" }), {
+              status: 200,
+            });
+          }
+          if (url.includes("/api/mcp/events")) {
+            sse = createPushableSseStream();
+            return sse.response;
+          }
+          if (url.endsWith("/api/mcp/send")) {
+            seenInits.push(init);
+            return new Response(JSON.stringify({ ok: true }), { status: 200 });
+          }
+          return new Response("not found", { status: 404 });
+        });
+
+      const transport = new RemoteClientTransport({ baseUrl, fetchFn }, config);
+      await transport.start();
+
+      const controller = new AbortController();
+      await transport.send(
+        { jsonrpc: "2.0", method: "notifications/initialized" },
+        { requestSignal: controller.signal },
+      );
+      expect(seenInits[0]?.signal).toBe(controller.signal);
+
+      // No signal supplied (stdio/SSE, where the SDK never creates one) must
+      // leave the fetch unaborted rather than passing `undefined` through as a
+      // deliberate value.
+      await transport.send({
+        jsonrpc: "2.0",
+        method: "notifications/roots/list_changed",
+      });
+      expect(seenInits[1]).not.toHaveProperty("signal");
+
+      await transport.close();
+    });
+
+    it("aborting the requestSignal rejects the in-flight send and releases its response wait", async () => {
+      let sse = createPushableSseStream();
+      const fetchFn = vi
+        .fn<typeof fetch>()
+        .mockImplementation(async (input, init) => {
+          const url = String(input);
+          if (url.endsWith("/api/mcp/connect")) {
+            return new Response(JSON.stringify({ sessionId: "abc" }), {
+              status: 200,
+            });
+          }
+          if (url.includes("/api/mcp/events")) {
+            sse = createPushableSseStream();
+            return sse.response;
+          }
+          if (url.endsWith("/api/mcp/send")) {
+            // The backend holds this route open for the whole call, so model it
+            // as a fetch that only settles when the caller aborts.
+            return new Promise<Response>((_resolve, reject) => {
+              init?.signal?.addEventListener("abort", () =>
+                reject(new DOMException("Aborted", "AbortError")),
+              );
+            });
+          }
+          return new Response("not found", { status: 404 });
+        });
+
+      const transport = new RemoteClientTransport({ baseUrl, fetchFn }, config);
+      await transport.start();
+
+      const controller = new AbortController();
+      const sent = transport.send(
+        { jsonrpc: "2.0", id: 7, method: "tools/call" },
+        { requestSignal: controller.signal },
+      );
+      const settled = vi.fn();
+      void sent.then(settled, settled);
+
+      controller.abort();
+      await expect(sent).rejects.toThrow(/Aborted/);
+      // The response wait registered for id 7 must have been cancelled too —
+      // otherwise this send would hang past the abort waiting for a JSON-RPC
+      // response the aborted request can never produce.
+      expect(settled).toHaveBeenCalledTimes(1);
+
+      await transport.close();
+    });
+  });
 });

clients/web/src/test/integration/mcp/inspectorClient-modern-era.test.ts

@@ -437,6 +437,63 @@ describe("modern-era negotiation (2026-07-28)", () => {
     expect(result.result!.isError).toBeFalsy();
   });
 
+  // #2140: on the 2026-07-28 era, closing the request's own response stream IS
+  // the cancellation signal for Streamable HTTP; `notifications/cancelled` is
+  // the stdio mechanism and the spec says it is neither required nor expected
+  // here. The SDK implements that fork off `transport.hasPerRequestStream` —
+  // but every Inspector connection is wrapped in `MessageTrackingTransport`,
+  // which was not forwarding it, so all three clients POSTed the notification
+  // and a spec-compliant server acknowledged it `202` and kept running.
+  //
+  // This asserts at the far end, on the server's own request signal: nothing on
+  // the client side distinguishes a cancel that reached the server from one
+  // that was dropped, which is exactly how this shipped.
+  it("cancels by aborting the request's stream, which the server observes", async () => {
+    let sawAbort!: (value: boolean) => void;
+    const aborted = new Promise<boolean>((resolve) => {
+      sawAbort = resolve;
+    });
+    let started!: () => void;
+    const running = new Promise<void>((resolve) => {
+      started = resolve;
+    });
+
+    const slowServer = createTestServerHttp({
+      serverInfo: createTestServerInfo("cancel-era-test", "1.0.0"),
+      tools: [
+        {
+          name: "slow_task",
+          description: "Runs until the client cancels it",
+          handler: async (_params, _context, extra) => {
+            extra?.signal?.addEventListener("abort", () => sawAbort(true), {
+              once: true,
+            });
+            started();
+            // Never settles on its own, so a regression is a timeout rather
+            // than a pass.
+            return new Promise(() => {});
+          },
+        },
+      ],
+      modern: {},
+    });
+    await slowServer.start();
+    server = slowServer;
+
+    const connected = await connectWithEra(slowServer.url, "modern");
+    const { tools } = await connected.listTools();
+    const tool = tools.find((t) => t.name === "slow_task");
+
+    // Hold the rejection immediately so it is never seen as unhandled while we
+    // wait on the server side.
+    const settled = connected.callTool(tool!, {}).catch((err: unknown) => err);
+    await running;
+    expect(connected.cancelToolCall()).toBe(true);
+
+    await expect(settled).resolves.toBeInstanceOf(ToolCallCancelledError);
+    await expect(aborted).resolves.toBe(true);
+  }, 30_000);
+
   it("cancels an in-flight MRTR call while its embedded request is pending", async () => {
     const started = await startMrtrServer(createMrtrTool());
     const connected = await connectWithEra(started.url, "modern");

clients/web/src/test/integration/mcp/remote/cancel-per-request-stream.test.ts

@@ -0,0 +1,199 @@
+/**
+ * #2140 — the Cancel button must send the transport's cancellation signal.
+ *
+ * On a 2026-07-28 Streamable HTTP connection the spec makes *closing the
+ * request's SSE response stream* the cancellation signal, and says a
+ * `notifications/cancelled` is neither required nor expected. The SDK
+ * implements that fork itself: `Protocol.request` aborts a per-request
+ * `AbortController` — handed to the transport as
+ * `TransportSendOptions.requestSignal` — when the connection is modern **and**
+ * the transport advertises `hasPerRequestStream`; otherwise it POSTs the
+ * notification.
+ *
+ * The web client lost it at the proxy boundary. The real upstream transport
+ * lives on the Node backend, and the browser's `RemoteClientTransport` neither
+ * advertised the flag nor applied a `requestSignal` to its
+ * `POST /api/mcp/send` — so every web cancel took the stdio branch, the server
+ * answered the notification `202 Accepted` and dropped it, and the tool kept
+ * running to completion. Only a manual disconnect actually cancelled anything.
+ *
+ * This drives the whole chain the browser drives — InspectorClient ->
+ * RemoteClientTransport -> the Hono backend -> a real upstream Streamable HTTP
+ * transport -> a modern test server — and asserts at the far end, on the
+ * server's own request abort signal. Nothing shallower reaches it: each of the
+ * three hops can drop the signal on its own, and a unit test of any one of them
+ * passes while the chain stays broken.
+ */
+
+import { describe, it, expect, afterEach } from "vitest";
+import { serve } from "@hono/node-server";
+import type { ServerType } from "@hono/node-server";
+import { InspectorClient } from "@inspector/core/mcp/inspectorClient.js";
+import { createRemoteTransport } from "@inspector/core/mcp/remote/createRemoteTransport.js";
+import { createRemoteApp } from "@inspector/core/mcp/remote/node/server.js";
+import { ToolCallCancelledError } from "@inspector/core/mcp/toolCallCancelledError.js";
+import { eraToVersionNegotiation } from "@inspector/core/mcp/types.js";
+import {
+  createTestServerHttp,
+  createTestServerInfo,
+  type TestServerHttp,
+  type ToolDefinition,
+} from "@modelcontextprotocol/inspector-test-server";
+
+function deferred<T>(): { promise: Promise<T>; resolve: (value: T) => void } {
+  let resolve!: (value: T) => void;
+  const promise = new Promise<T>((r) => {
+    resolve = r;
+  });
+  return { promise, resolve };
+}
+
+/**
+ * A tool that never returns on its own. It reports when it started, and whether
+ * the server-side request signal — which the SDK aborts when the client closes
+ * this request's response stream — ever fired.
+ */
+function createSlowTool(): {
+  tool: ToolDefinition;
+  started: Promise<void>;
+  aborted: Promise<boolean>;
+} {
+  const started = deferred<void>();
+  const aborted = deferred<boolean>();
+  const tool: ToolDefinition = {
+    name: "slow_task",
+    description: "Runs until the client cancels it",
+    handler: async (_params, _context, extra) => {
+      extra?.signal?.addEventListener("abort", () => aborted.resolve(true), {
+        once: true,
+      });
+      started.resolve();
+      // Never settles: this call ends by cancellation or not at all, so a
+      // regression shows up as the test's own timeout rather than as a pass.
+      return new Promise(() => {});
+    },
+  };
+  return { tool, started: started.promise, aborted: aborted.promise };
+}
+
+async function startRemoteBackend(): Promise<{
+  baseUrl: string;
+  server: ServerType;
+  authToken: string;
+}> {
+  const { app, authToken } = createRemoteApp({
+    initialConfig: { defaultEnvironment: {} },
+  });
+  return new Promise((resolve, reject) => {
+    const server = serve(
+      { fetch: app.fetch, port: 0, hostname: "127.0.0.1" },
+      (info) => {
+        const port =
+          info && typeof info === "object" && "port" in info
+            ? (info as { port: number }).port
+            : 0;
+        resolve({ baseUrl: `http://127.0.0.1:${port}`, server, authToken });
+      },
+    );
+    server.on("error", reject);
+  });
+}
+
+describe("web-client cancellation over a per-request stream (#2140)", () => {
+  let client: InspectorClient | null = null;
+  let backend: ServerType | null = null;
+  let upstream: TestServerHttp | null = null;
+
+  afterEach(async () => {
+    if (client) {
+      try {
+        await client.disconnect();
+      } catch {
+        // Ignore disconnect errors.
+      }
+      client = null;
+    }
+    if (backend) {
+      const server = backend;
+      backend = null;
+      await new Promise<void>((resolve, reject) => {
+        server.close((err) => (err ? reject(err) : resolve()));
+      });
+    }
+    if (upstream) {
+      try {
+        await upstream.stop();
+      } catch {
+        // Ignore stop errors.
+      }
+      upstream = null;
+    }
+  });
+
+  it("aborts the upstream request stream, which the server observes as cancellation", async () => {
+    const slow = createSlowTool();
+    upstream = createTestServerHttp({
+      serverInfo: createTestServerInfo("cancel-test", "1.0.0"),
+      tools: [slow.tool],
+      modern: {},
+    });
+    await upstream.start();
+
+    const { baseUrl, server, authToken } = await startRemoteBackend();
+    backend = server;
+
+    const connected = new InspectorClient(
+      { type: "streamable-http", url: upstream.url },
+      {
+        environment: {
+          transport: createRemoteTransport({ baseUrl, authToken }),
+        },
+        versionNegotiation: eraToVersionNegotiation("modern"),
+      },
+    );
+    await connected.connect();
+    client = connected;
+    expect(connected.getProtocolEra()).toBe("modern");
+
+    const { tools } = await connected.listTools();
+    const tool = tools.find((t) => t.name === "slow_task");
+    expect(tool).toBeDefined();
+
+    // Hold the rejection from the first turn so the runner never sees it as
+    // unhandled while we wait on the server side.
+    const settled = connected.callTool(tool!, {}).catch((err: unknown) => err);
+
+    await slow.started;
+    expect(connected.cancelToolCall()).toBe(true);
+
+    // The user-facing contract is unchanged: a deliberate cancel still surfaces
+    // as ToolCallCancelledError, whichever wire signal carried it.
+    await expect(settled).resolves.toBeInstanceOf(ToolCallCancelledError);
+
+    // The assertion that matters. Before the fix the server saw only a
+    // `notifications/cancelled` it was free to ignore (and did), its request
+    // signal never fired, and this promise never resolved.
+    await expect(slow.aborted).resolves.toBe(true);
+  }, 30_000);
+
+  it("advertises the per-request stream only for streamable-http", () => {
+    // The flag is what the SDK forks on, and the factory is where a server's
+    // configured type reaches the transport. stdio and SSE multiplex every
+    // request over one shared channel: there is no per-request stream to close,
+    // and aborting anything would take the whole session down — so they must
+    // keep the `notifications/cancelled` mechanism.
+    const create = createRemoteTransport({ baseUrl: "http://unused.example" });
+    expect(
+      create({ type: "stdio", command: "echo", args: [] }, {}).transport
+        .hasPerRequestStream,
+    ).toBe(false);
+    expect(
+      create({ type: "sse", url: "http://unused.example/sse" }, {}).transport
+        .hasPerRequestStream,
+    ).toBe(false);
+    expect(
+      create({ type: "streamable-http", url: "http://unused.example/mcp" }, {})
+        .transport.hasPerRequestStream,
+    ).toBe(true);
+  });
+});