refactor(mcp): make MCP test reuse the shared Box session (near-instant tests) (#2304)

* refactor(mcp): make MCP test reuse the shared Box session instead of a per-test session

Testing an MCP server (config-page "test" button) previously spun up a fresh
isolated mcp-test-<uuid> Box session every time: cold-start the container, run
the dependency bootstrap, probe, then tear the whole session down. That is slow
(tens of seconds) and, on an already-hosted server, wasteful — the server is
already running in the shared session.

Change the test to reuse the shared session / live process:

- _build_box_session_id: transient tests now use mcp-shared, the same Box
  session as live servers, so a test reuses the running container (and, for an
  existing server, its live managed process) instead of a cold per-test session.
- cleanup_session: a transient test no longer deletes the whole session (which
  under the shared model would kill every other MCP server in the container). It
  stops only its own process_id, exactly like a live server. Isolation is now at
  the process level (distinct process_id per server/test), not the session level.
- test_mcp_server (persisted server): reuse the live connection with a real
  list_tools refresh/probe; only fall back to a full start() when there is no
  live connection to probe or the refresh fails, instead of an ERROR->start()
  rebuild.

Trade-off: a failing test now shares the container with live servers rather than
a throwaway session. Accepted deliberately in favour of near-instant tests;
process-level isolation keeps a test from stopping another server's process.

* chore(deps): pin langbot-plugin 0.4.9 for the nsjail RLIMIT_AS node/npx MCP fix

---------

Co-authored-by: dadachann <185672915+dadachann@users.noreply.github.com>

pyproject.toml

@@ -70,7 +70,7 @@ dependencies = [
     "chromadb>=1.0.0,<2.0.0",
     "qdrant-client (>=1.15.1,<2.0.0)",
     "pyseekdb==1.1.0.post3",
-    "langbot-plugin==0.4.8",
+    "langbot-plugin==0.4.9",
     "asyncpg>=0.30.0",
     "line-bot-sdk>=3.19.0",
     "matrix-nio>=0.25.2",

src/langbot/pkg/api/http/service/mcp.py

@@ -188,10 +188,22 @@ class MCPService:
             persisted_session = runtime_mcp_session
 
             async def _refresh_and_report() -> None:
-                if persisted_session.status == MCPSessionStatus.ERROR:
+                # Testing a persisted server should REUSE its live shared-session
+                # process, not rebuild it. Try a lightweight refresh (a real
+                # list_tools probe over the existing connection) first; only fall
+                # back to a full start() when the session has no live connection
+                # to probe (never connected, or the process is actually gone).
+                needs_start = persisted_session.status == MCPSessionStatus.ERROR or persisted_session.session is None
+                if needs_start:
                     await persisted_session.start()
                 else:
-                    await persisted_session.refresh()
+                    try:
+                        await persisted_session.refresh()
+                    except Exception:
+                        # The live connection was stale/dropped: reconnect once
+                        # (reusing the live managed process where possible) and
+                        # re-probe, instead of reporting a false failure.
+                        await persisted_session.start()
                 # Surface the discovered tools so the config page can render them
                 # even for an already-hosted server.
                 ctx.metadata['runtime_info'] = persisted_session.get_runtime_info_dict()

src/langbot/pkg/provider/tools/loaders/mcp.py

@@ -986,11 +986,14 @@ class RuntimeMCPSession:
         return self._box_stdio_runtime.uses_box_stdio()
 
     def _build_box_session_id(self) -> str:
-        # Transient test sessions get their own isolated Box session so a
+        # Both live servers and transient config-page tests share ONE Box
-        # failing/short-lived test can never disturb the shared session that
+        # session ('mcp-shared'). A test therefore reuses the already-running
-        # hosts live, already-connected MCP servers.
+        # container (and, for an existing server, its live managed process)
-        if self.is_transient:
+        # instead of paying a full per-test session cold-start + dependency
-            return f'mcp-test-{self.server_uuid}'
+        # bootstrap. Isolation between a test and the live servers is provided
+        # at the *process* level: each server/test has its own process_id and a
+        # test only ever stops its own process_id (see cleanup_session), so it
+        # never disturbs another server's process or the shared session itself.
         return 'mcp-shared'
 
     def _rewrite_path(self, path: str, host_path: str | None) -> str:

src/langbot/pkg/provider/tools/loaders/mcp_stdio.py

@@ -370,16 +370,20 @@ class BoxStdioSessionRuntime:
 
         workspace = self._build_workspace(host_path=None)
 
-        # Transient test sessions own their isolated Box session, so tear the
+        # Transient config-page tests now share the same 'mcp-shared' Box
-        # whole session down rather than leaking it. This cannot affect live
+        # session as live servers, so we must NOT tear the session down here —
-        # servers because they live in the separate shared session.
+        # that would kill every other MCP server in the container. A test is
+        # isolated at the process level: it ran under its own process_id, so we
+        # stop only that process, exactly like a live server does below. The
+        # shared session and all other servers' live processes are untouched.
+        # (Staged per-test workspace files are still cleaned up.)
         if getattr(self.owner, 'is_transient', False):
             try:
-                await workspace.cleanup()
+                await workspace.stop_managed_process(self.process_id)
             except Exception as exc:
                 self.ap.logger.warning(
-                    f'MCP server {self.server_name}: failed to delete transient test session '
+                    f'MCP server {self.server_name}: failed to stop transient test process '
-                    f'{self.owner._build_box_session_id()}: {type(exc).__name__}: {exc}'
+                    f'process_id={self.process_id}: {type(exc).__name__}: {exc}'
                 )
             await self._cleanup_staged_workspace()
             return

tests/unit_tests/provider/test_mcp_box_integration.py

@@ -639,10 +639,13 @@ class TestGetRuntimeInfoDict:
         assert info['box_session_id'] == 'mcp-shared'
         assert info['box_enabled'] is True
 
-    def test_transient_test_session_is_isolated_from_shared(self, mcp_module):
+    def test_transient_test_shares_session_but_isolated_by_process(self, mcp_module):
-        """A transient test session (config-page "test", no persisted UUID)
+        """A transient config-page "test" now shares the same 'mcp-shared' Box
-        must NOT share the live "mcp-shared" Box session. Regression: a failing
+        session as live servers (so a test reuses the running container / live
-        test churned the shared session and tore down healthy live servers."""
+        process instead of a cold per-test session bootstrap). Isolation is at
+        the PROCESS level: the test runs under its own process_id and only ever
+        stops that process_id, so it cannot disturb another server's live
+        process or the shared session itself."""
         ap = _make_ap()
         ap.box_service.available = True
         transient = _make_session(
@@ -670,10 +673,12 @@ class TestGetRuntimeInfoDict:
         )
         assert transient.is_transient is True
         assert live.is_transient is False
-        # Isolated session id for the test, shared for the live server.
+        # Both share ONE Box session ...
-        assert transient._build_box_session_id() == 'mcp-test-gen-uuid-123'
+        assert transient._build_box_session_id() == 'mcp-shared'
         assert live._build_box_session_id() == 'mcp-shared'
-        assert transient._build_box_session_id() != live._build_box_session_id()
+        assert transient._build_box_session_id() == live._build_box_session_id()
+        # ... but are isolated by distinct process_ids within that session.
+        assert transient._box_stdio_runtime.process_id != live._box_stdio_runtime.process_id
 
     def test_stdio_session_refuses_when_box_unavailable(self, mcp_module):
         """Policy: when Box is configured but unavailable (disabled in config

uv.lock

@@ -2123,7 +2123,7 @@ requires-dist = [
     { name = "ebooklib", specifier = ">=0.18" },
     { name = "gewechat-client", specifier = ">=0.1.5" },
     { name = "html2text", specifier = ">=2024.2.26" },
-    { name = "langbot-plugin", specifier = "==0.4.8" },
+    { name = "langbot-plugin", specifier = "==0.4.9" },
     { name = "langchain", specifier = ">=1.3.9" },
     { name = "langchain-core", specifier = ">=1.3.3" },
     { name = "langchain-text-splitters", specifier = ">=1.1.2" },
@@ -2187,7 +2187,7 @@ dev = [
 
 [[package]]
 name = "langbot-plugin"
-version = "0.4.8"
+version = "0.4.9"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "aiofiles" },
@@ -2208,9 +2208,9 @@ dependencies = [
     { name = "watchdog" },
     { name = "websockets" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/7f/f5/1b417a48d329f959a261eaf9a37b26197f18081d2b6f0ef3e4bbab687b81/langbot_plugin-0.4.8.tar.gz", hash = "sha256:e9614be248acf352acb9430d9e154e0abae981575de16812f68282c92a4e42d8", size = 334076, upload-time = "2026-07-03T04:15:28.233Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/54/52/f85939d3be929696dc5f9f3369f18bd6652981a8985c6e6f0dfe48e04f75/langbot_plugin-0.4.9.tar.gz", hash = "sha256:1cc2882f1be96cd52be6c392c3925e44cf785c97c10f82a321339636da318c6e", size = 334682, upload-time = "2026-07-03T04:57:42.813Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/40/21/46bfa911ab27d6856a3b4363805c4bb8a657500cbc12e1e719d0b3b349dc/langbot_plugin-0.4.8-py3-none-any.whl", hash = "sha256:87c12d9d5579b0ab4dc43c9e559b4d15d7387db8c8b52d476cf27a3de74aedf5", size = 221244, upload-time = "2026-07-03T04:15:27.087Z" },
+    { url = "https://files.pythonhosted.org/packages/47/12/1cf535377a81cf01607bbee5e3f4745a8c9f6dbb561887467001a182a533/langbot_plugin-0.4.9-py3-none-any.whl", hash = "sha256:693225a089ba38bc8d0b58d9e0fdf07b027b5be6b5e858cf7b4e56d095b68ba5", size = 221722, upload-time = "2026-07-03T04:57:41.578Z" },
 ]
 
 [[package]]