diff --git a/docs/mcp-catalog-evaluation.md b/docs/mcp-catalog-evaluation.md
index eff14206d8..1e2713a8f1 100644
--- a/docs/mcp-catalog-evaluation.md
+++ b/docs/mcp-catalog-evaluation.md
@@ -39,6 +39,12 @@ and a 192-task ceiling. Runs are sequential. Unsupported clients, resource stops
 failed startup, missing usage, and unverified cleanup remain failed measurements.
 They do not become zero-cost successes.
 
+Trials disable account apps and plugins explicitly. A fresh `CODEX_HOME` alone
+does not prevent account-connected tools from appearing in the client executor
+catalog. For a deliberate mixed-catalog investigation, run `agent-trial.py` with
+`--include-account-tools`; keep those results separate from REA-only trials.
+The settings receipt records the same feature policy written to Codex config.
+
 Each repetition compares no server, all REA tools, and a client-selected subset
 (`binary_session` by default), in the supported client-code host mode. Alternate
 repetitions reverse ordering. All variants receive the same developer instructions
diff --git a/src/catalogDiscovery.ts b/src/catalogDiscovery.ts
index 1b4b31a64e..6a5724f4ca 100644
--- a/src/catalogDiscovery.ts
+++ b/src/catalogDiscovery.ts
@@ -12,7 +12,13 @@ export const HOT_TOOL_NAMES: ReadonlySet<string> = new Set([
 
 // One page activates at most 20 schemas: callers can page a broad search instead
 // of inadvertently expanding the entire catalog into their client's context.
-const pageSize = z.int().min(1).max(20).default(3);
+const MAX_DISCOVERY_TOOLS = 20;
+const DEFAULT_DISCOVERY_PAGE_SIZE = 3;
+const pageSize = z
+  .int()
+  .min(1)
+  .max(MAX_DISCOVERY_TOOLS)
+  .default(DEFAULT_DISCOVERY_PAGE_SIZE);
 const inputSchema = z
   .strictObject({
     query: z
@@ -25,12 +31,14 @@ const inputSchema = z
     names: z
       .array(z.string().min(1))
       .min(1)
-      .max(20)
+      .max(MAX_DISCOVERY_TOOLS)
       .optional()
-      .describe("Exact tool names to load instead of searching."),
+      .describe(
+        `At most ${MAX_DISCOVERY_TOOLS} exact tool names to load instead of searching.`,
+      ),
     offset: z.int().min(0).default(0),
     limit: pageSize.describe(
-      "Search page size; does not apply to exact names.",
+      `Search page size 1..${MAX_DISCOVERY_TOOLS} (default ${DEFAULT_DISCOVERY_PAGE_SIZE}); does not apply to exact names.`,
     ),
   })
   .refine(
diff --git a/tests/support/analysis-efficiency/agent-trial.py b/tests/support/analysis-efficiency/agent-trial.py
index f5db80f9b9..8308c95c93 100644
--- a/tests/support/analysis-efficiency/agent-trial.py
+++ b/tests/support/analysis-efficiency/agent-trial.py
@@ -37,6 +37,8 @@ def main():
     parser.add_argument('--profile-parent', type=Path,
                         help='External disposable-profile parent (default: cache beside the source Codex home)')
     parser.add_argument('--authority', choices=['static', 'browser'], default='static')
+    parser.add_argument('--include-account-tools', action='store_true',
+                        help='Allow account apps/plugins; default isolates REA from their catalogs')
     parser.add_argument('--capture-tool-context', action='store_true',
                         help='Retain owned tool/usage items; excludes account/ambient state')
     parser.add_argument('--guide-enabled-tools', action='store_true',
@@ -159,14 +161,20 @@ def main():
             'or delegate. Reply with exactly the text requested by the user.'
         )
     quote = json.dumps
+    # A disposable CODEX_HOME does not disable account-connected apps. Keep
+    # the requested feature policy identical in the config and its receipt.
+    features = {
+        'multi_agent': False, 'shell_snapshot': False, 'shell_tool': False,
+        'unified_exec': False, 'code_mode_host': args.mode == 'client-code',
+        'apps': args.include_account_tools, 'plugins': args.include_account_tools,
+    }
     config = '\n'.join([
         'approval_policy = "never"', 'sandbox_mode = "read-only"',
         'web_search = "disabled"', 'model = ' + quote(args.model),
         'model_reasoning_effort = ' + quote(args.reasoning),
         'developer_instructions = ' + quote(instruction),
-        '[features]', 'multi_agent = false', 'shell_snapshot = false',
-        'shell_tool = false', 'unified_exec = false',
-        'code_mode_host = ' + ('true' if args.mode == 'client-code' else 'false'),
+        '[features]',
+        *[name + ' = ' + str(enabled).lower() for name, enabled in features.items()],
         '[history]', 'persistence = "none"',
     ]) + '\n'
     if not args.no_mcp:
@@ -253,8 +261,7 @@ def main():
                 provider_env.get('PATH', os.environ.get('PATH', '')).encode()).hexdigest(),
             'provider_environment_policy_sha256': hashlib.sha256(
                 provider_policy_text.encode()).hexdigest(),
-            'features': {'shell_tool': False, 'unified_exec': False,
-                         'multi_agent': False, 'code_mode_host': args.mode == 'client-code'},
+            'features': features,
         }, indent=2) + '\n')
         invocation = [args.codex, '--no-daemon', 'exec', '--strict-config',
                       *([] if args.capture_tool_context else ['--ephemeral']),
diff --git a/tests/support/analysis-efficiency/test_codex_context.py b/tests/support/analysis-efficiency/test_codex_context.py
index 6b5c1350e7..2ebc6fb2a2 100644
--- a/tests/support/analysis-efficiency/test_codex_context.py
+++ b/tests/support/analysis-efficiency/test_codex_context.py
@@ -681,9 +681,12 @@ class ContextCaptureTests(unittest.TestCase):
         spec = importlib.util.spec_from_file_location('agent_trial', Path(__file__).with_name('agent-trial.py'))
         agent = importlib.util.module_from_spec(spec)
         spec.loader.exec_module(agent)
-        for control, drift, mode in [(False, False, 'direct'), (True, False, 'direct'),
-                                     (False, True, 'direct'), (False, False, 'client-code')]:
-            with self.subTest(control=control, drift=drift, mode=mode), tempfile.TemporaryDirectory() as directory:
+        for control, drift, mode, account_tools in [
+            (False, False, 'direct', False), (True, False, 'direct', False),
+            (False, True, 'direct', False), (False, False, 'client-code', False),
+            (False, False, 'client-code', True), (True, False, 'client-code', True),
+        ]:
+            with self.subTest(control=control, drift=drift, mode=mode, account_tools=account_tools), tempfile.TemporaryDirectory() as directory:
                 root = Path(directory)
                 runtime = root / 'runtime'
                 (runtime / 'scripts').mkdir(parents=True)
@@ -696,7 +699,8 @@ class ContextCaptureTests(unittest.TestCase):
                 output = runtime / '.cache/trial'
                 arguments = ['agent-trial.py', '--runtime-root', str(runtime), '--revision', 'fixture',
                              '--prompt', str(question), '--output', str(output), '--catalog-load', '--mode', mode,
-                             *(['--no-mcp'] if control else [])]
+                             *(['--no-mcp'] if control else []),
+                             *(['--include-account-tools'] if account_tools else [])]
                 observed = {}
 
                 def launch(_invocation, **options):
@@ -723,6 +727,12 @@ class ContextCaptureTests(unittest.TestCase):
                 self.assertEqual('required = true' in observed['config'], not control)
                 self.assertNotIn('Use the registered REA MCP server', observed['config'])
                 self.assertIn('code_mode_host = ' + ('true' if mode == 'client-code' else 'false'), observed['config'])
+                settings = json.loads((output / 'trial-settings.json').read_text())
+                for feature in ['apps', 'plugins']:
+                    self.assertIn(feature + ' = ' + str(account_tools).lower(), observed['config'])
+                    self.assertIs(settings['features'][feature], account_tools)
+                for feature, enabled in settings['features'].items():
+                    self.assertIn(feature + ' = ' + str(enabled).lower(), observed['config'])
                 self.assertNotIn('code_mode =', observed['config'])
                 self.assertNotIn('code_mode_only =', observed['config'])
                 self.assertTrue(json.loads((output / 'trial-settings.json').read_text())['workspace_isolated_from_runtime'])