refactor(compression): scope Codex-native compaction to the app-server runtime

Drop the Responses-API native compaction path and its opt-in umbrella
flag from the salvaged feature. On the Codex OAuth chat route Hermes
owns the message list and the summary compressor works (and stays
provider-portable — encrypted compaction items would lock the session
history to chatgpt.com and break /model switches and provider
fallback). On the app-server runtime (codex CLI/agent) the codex agent
owns the real thread context, so thread/compact/start is the only
mechanism that can actually shrink it (#36801) — that path is now the
default behavior for codex_app_server sessions, controlled by
compression.codex_app_server_auto (native|hermes|off), no umbrella
flag.

Removed: responses.compact() call path, codex_compaction_items replay/
persistence plumbing, codex_native_compaction + codex_responses_threshold
config keys, desktop settings fields, and their tests. Kept: everything
app-server (compact_thread(), compaction notifications, bookkeeping,
docs, tests) plus cache-busting keys for the surviving knobs.

cli-config.yaml.example

@@ -415,17 +415,6 @@ compression:
   # for the ChatGPT Codex OAuth route. Set false to opt back down to threshold.
   codex_gpt55_autoraise: true
 
-  # Codex-native compaction paths are opt-in (default: false) to preserve
-  # Hermes' existing auxiliary summarizer behavior for existing installs.
-  # When true, Codex OAuth uses Responses API compact and Codex app-server
-  # compaction uses the app-server thread compact API.
-  codex_native_compaction: false
-
-  # Codex OAuth / Responses API compaction trigger (default: 0.85 = 85%).
-  # Used only when codex_native_compaction is true. The recent tail still uses
-  # protect_last_n below.
-  codex_responses_threshold: 0.85
-  
   # Fraction of the threshold to preserve as recent tail (default: 0.20 = 20%)
   # e.g. 20% of 50% threshold = 10% of total context kept as recent messages.
   # Summary output is separately capped at 12K tokens (Gemini output limit).
@@ -437,8 +426,9 @@ compression:
   # compression of older turns.
   protect_last_n: 20
 
-  # Codex app-server auto-compaction mode:
-  # Used only when codex_native_compaction is true.
+  # Codex app-server (codex CLI runtime) thread-compaction mode. The codex
+  # agent owns the real thread context on this runtime, so Hermes' summarizer
+  # cannot shrink it — compaction goes through the app server instead.
   #   native = let Codex decide when to compact its own thread (default)
   #   hermes = let Hermes threshold trigger Codex thread/compact/start
   #   off    = Hermes will not auto-trigger compaction; Codex may still compact natively

gateway/run.py

@@ -15321,6 +15321,8 @@ class GatewayRunner(GatewayAuthorizationMixin, GatewayKanbanWatchersMixin, Gatew
         ("model", "max_tokens"),
         ("compression", "enabled"),
         ("compression", "threshold"),
+        ("compression", "codex_gpt55_autoraise"),
+        ("compression", "codex_app_server_auto"),
         ("compression", "target_ratio"),
         ("compression", "protect_last_n"),
         ("agent", "disabled_toolsets"),

tests/gateway/test_agent_cache.py

@@ -229,8 +229,6 @@ class TestExtractCacheBustingConfig:
                     "enabled": False,
                     "threshold": 0.6,
                     "codex_gpt55_autoraise": False,
-                    "codex_native_compaction": True,
-                    "codex_responses_threshold": 0.85,
                     "target_ratio": 0.3,
                     "protect_last_n": 25,
                     "codex_app_server_auto": "hermes",
@@ -241,8 +239,6 @@ class TestExtractCacheBustingConfig:
         assert out["compression.enabled"] is False
         assert out["compression.threshold"] == 0.6
         assert out["compression.codex_gpt55_autoraise"] is False
-        assert out["compression.codex_native_compaction"] is True
-        assert out["compression.codex_responses_threshold"] == 0.85
         assert out["compression.target_ratio"] == 0.3
         assert out["compression.protect_last_n"] == 25
         assert out["compression.codex_app_server_auto"] == "hermes"

tests/hermes_cli/test_config.py

@@ -1782,52 +1782,28 @@ class TestConfigNormalizationDoesNotOverwriteUserValues:
         assert _explicit_config_paths({"memory": {}, "display": {}}) == set()
 
 
-class TestCodexNativeCompactionConfig:
-    """Codex-native compaction stays opt-in without mutating existing configs."""
+class TestCodexAppServerAutoConfig:
+    """codex_app_server_auto ships a default and survives migration untouched."""
 
     def _write(self, tmp_path, body):
         (tmp_path / "config.yaml").write_text(body, encoding="utf-8")
 
-    def test_default_config_keeps_codex_native_compaction_opt_in(self):
-        assert DEFAULT_CONFIG["compression"]["codex_native_compaction"] is False
+    def test_default_config_has_native_mode(self):
+        assert DEFAULT_CONFIG["compression"]["codex_app_server_auto"] == "native"
         assert DEFAULT_CONFIG["compression"]["codex_gpt55_autoraise"] is True
 
-    def test_migration_does_not_write_codex_native_compaction_default(self, tmp_path):
+    def test_preserves_existing_codex_app_server_auto_value(self, tmp_path):
         with patch.dict(os.environ, {"HERMES_HOME": str(tmp_path)}):
             self._write(
                 tmp_path,
                 "_config_version: 31\n"
                 "compression:\n"
-                "  threshold: 0.5\n"
-                "  codex_gpt55_autoraise: false\n",
+                "  codex_app_server_auto: hermes\n",
             )
 
-            result = migrate_config(interactive=False, quiet=True)
+            migrate_config(interactive=False, quiet=True)
 
             raw = yaml.safe_load((tmp_path / "config.yaml").read_text())
-            assert raw["compression"]["codex_gpt55_autoraise"] is False
-            assert "codex_native_compaction" not in raw["compression"]
-            assert raw["compression"]["threshold"] == 0.5
-            assert raw["_config_version"] == DEFAULT_CONFIG["_config_version"]
-            assert (
-                "compression.codex_native_compaction=false"
-                not in result["config_added"]
-            )
+            assert raw["compression"]["codex_app_server_auto"] == "hermes"
 
-    def test_preserves_existing_codex_native_compaction_value(self, tmp_path):
-        with patch.dict(os.environ, {"HERMES_HOME": str(tmp_path)}):
-            self._write(
-                tmp_path,
-                "_config_version: 31\n"
-                "compression:\n"
-                "  codex_native_compaction: true\n",
-            )
 
-            result = migrate_config(interactive=False, quiet=True)
-
-            raw = yaml.safe_load((tmp_path / "config.yaml").read_text())
-            assert raw["compression"]["codex_native_compaction"] is True
-            assert (
-                "compression.codex_native_compaction=false"
-                not in result["config_added"]
-            )

tests/run_agent/test_codex_app_server_compaction.py

@@ -25,10 +25,8 @@ class DummyAgent:
         result,
         *,
         auto_compaction="native",
-        codex_native_compaction=True,
     ):
         self.api_mode = "codex_app_server"
-        self.codex_native_compaction_enabled = codex_native_compaction
         self.codex_app_server_auto_compaction = auto_compaction
         self.session_id = "hermes-session-1"
         self.platform = "cli"