From 232254b6d90b246dd5e07151ecada24021af6085 Mon Sep 17 00:00:00 2001
From: Haoqian Li <haoqian.li@finalroundai.com>
Date: Thu, 25 Jun 2026 01:20:07 +0800
Subject: [PATCH] fix(mcp): arm liveness watchdog in local proxy

---
 __tests__/mcp-daemon.test.ts | 12 +++++++-----
 src/mcp/proxy.ts             |  6 ++++++
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/__tests__/mcp-daemon.test.ts b/__tests__/mcp-daemon.test.ts
index c00d528f6..c63a5d929 100644
--- a/__tests__/mcp-daemon.test.ts
+++ b/__tests__/mcp-daemon.test.ts
@@ -193,7 +193,10 @@ describe('Shared MCP daemon (issue #411)', () => {
   });
 
   it('two invocations share ONE detached daemon; both attach as proxies', async () => {
-    const env = { CODEGRAPH_DAEMON_IDLE_TIMEOUT_MS: '15000' };
+    const env = {
+      CODEGRAPH_DAEMON_IDLE_TIMEOUT_MS: '15000',
+      CODEGRAPH_MCP_DEBUG: '1',
+    };
 
     const first = spawnServer(tempDir, env);
     servers.push(first);
@@ -203,6 +206,9 @@ describe('Shared MCP daemon (issue #411)', () => {
 
     // The launcher is a PROXY (not the daemon itself) — that's the detach fix.
     await waitFor(() => first.stderr.some((l) => l.includes('Attached to shared daemon')), 8000);
+    // The local-handshake proxy also needs the main-thread liveness watchdog
+    // (#943), matching direct and detached-daemon mode.
+    await waitFor(() => first.stderr.some((l) => l.includes('[CodeGraph watchdog] armed')), 8000);
 
     // A detached daemon came up and recorded itself.
     await waitFor(() => fs.existsSync(path.join(realRoot, '.codegraph', 'daemon.pid')), 8000);
@@ -416,10 +422,6 @@ describe('Shared MCP daemon (issue #411)', () => {
     await waitFor(() => (readLockPid(realRoot) ?? 0) > 0, 8000);
     const daemonPid = readLockPid(realRoot)!;
 
-    // A warm call goes through the daemon.
-    sendMessage(server.child, { jsonrpc: '2.0', id: 2, method: 'tools/call', params: { name: 'codegraph_status', arguments: {} } });
-    await waitFor(() => findResponse(server.stdout, 2), 10000);
-
     // Kill the daemon out from under the live proxy.
     process.kill(daemonPid, 'SIGTERM');
     expect(await waitProcessExit(daemonPid, 8000)).toBe(true);
diff --git a/src/mcp/proxy.ts b/src/mcp/proxy.ts
index b7b538dd4..346d76970 100644
--- a/src/mcp/proxy.ts
+++ b/src/mcp/proxy.ts
@@ -29,6 +29,7 @@ import { SERVER_INFO, PROTOCOL_VERSION } from './session';
 import { SERVER_INSTRUCTIONS } from './server-instructions';
 import { getStaticTools } from './tools';
 import { getTelemetry, ClientInfo } from '../telemetry';
+import { installMainThreadWatchdog, WatchdogHandle } from './liveness-watchdog';
 import type { MCPEngine } from './engine';
 
 /** Default poll cadence for the PPID watchdog (same as the direct server). */
@@ -202,6 +203,10 @@ export interface LocalHandshakeDeps {
  * never costs the old fall-back-to-direct robustness.
  */
 export async function runLocalHandshakeProxy(deps: LocalHandshakeDeps): Promise<void> {
+  // The proxy is long-lived and can serve fallback tool calls in-process. Match
+  // direct/daemon mode by killing this launcher if its main thread wedges, so an
+  // MCP host retry cannot accumulate abandoned `serve --mcp` wrapper processes.
+  const livenessWatchdog: WatchdogHandle | null = installMainThreadWatchdog();
   let daemonStatus: 'connecting' | 'ready' | 'failed' = 'connecting';
   let daemonSocket: net.Socket | null = null;
   let clientInitId: unknown = undefined;   // suppress the daemon's reply to the forwarded initialize
@@ -232,6 +237,7 @@ export async function runLocalHandshakeProxy(deps: LocalHandshakeDeps): Promise<
   };
   const shutdown = (): void => {
     if (shuttingDown) return; shuttingDown = true;
+    try { livenessWatchdog?.stop(); } catch { /* ignore */ }
     try { daemonSocket?.destroy(); } catch { /* ignore */ }
     try { engine?.stop(); } catch { /* ignore */ }
     process.exit(0);