refactor(context): trim replay cap plumbing

nanobot/agent/loop.py

@@ -205,7 +205,6 @@ class AgentLoop:
         timezone: str | None = None,
         session_ttl_minutes: int = 0,
         consolidation_ratio: float = 0.5,
-        max_messages: int = 0,
         hooks: list[AgentHook] | None = None,
         unified_session: bool = False,
         disabled_skills: list[str] | None = None,
@@ -296,12 +295,7 @@ class AgentLoop:
             llm_wall_timeout_for_session=lambda sk: runner_wall_llm_timeout_s(self.sessions, sk),
         )
         self._unified_session = unified_session
-        self._explicit_max_messages = max_messages > 0
-        self._max_messages = (
-            max_messages
-            if self._explicit_max_messages
-            else replay_max_messages_for_context(self.context_window_tokens)
-        )
+        self._max_messages = replay_max_messages_for_context(self.context_window_tokens)
         self._running = False
         self._mcp_servers = mcp_servers or {}
         self._mcp_stacks: dict[str, AsyncExitStack] = {}
@@ -444,8 +438,7 @@ class AgentLoop:
         logger.info("Runtime model switched for next turn: {} -> {}", old_model, model)
 
     def _sync_replay_max_messages(self) -> None:
-        if not self._explicit_max_messages:
-            self._max_messages = replay_max_messages_for_context(self.context_window_tokens)
+        self._max_messages = replay_max_messages_for_context(self.context_window_tokens)
 
     def _refresh_provider_snapshot(self) -> None:
         if self._provider_snapshot_loader is None:

nanobot/config/loader.py

@@ -156,12 +156,12 @@ def _migrate_config(data: dict) -> dict:
     agents = data.get("agents", {})
     defaults = agents.get("defaults", {}) if isinstance(agents, dict) else {}
     if isinstance(defaults, dict):
-        legacy_max_message_keys = [
-            key for key in ("maxMessages", "max_messages") if key in defaults
-        ]
-        if legacy_max_message_keys:
-            for key in legacy_max_message_keys:
-                defaults.pop(key, None)
+        had_legacy_max_messages = (
+            "maxMessages" in defaults or "max_messages" in defaults
+        )
+        defaults.pop("maxMessages", None)
+        defaults.pop("max_messages", None)
+        if had_legacy_max_messages:
             # TODO(next version): Remove this legacy cleanup branch; the schema
             # will silently ignore this field once the warning grace period ends.
             logger.warning(

nanobot/session/manager.py

@@ -29,7 +29,6 @@ from nanobot.utils.subagent_channel_display import scrub_subagent_announce_body
 FILE_MAX_MESSAGES = 2000
 MIN_REPLAY_MAX_MESSAGES = 120
 REPLAY_TOKENS_PER_MESSAGE = 100
-DEFAULT_REPLAY_MAX_MESSAGES = FILE_MAX_MESSAGES
 _MESSAGE_TIME_PREFIX_RE = re.compile(r"^\[Message Time: [^\]]+\]\n?")
 _LOCAL_IMAGE_BREADCRUMB_RE = re.compile(r"^\[image: (?:/|~)[^\]]+\]\s*$")
 _TOOL_CALL_ECHO_RE = re.compile(r'^\s*(?:generate_image|message)\([^)]*\)\s*$')
@@ -48,7 +47,7 @@ _FORK_VOLATILE_METADATA_KEYS = {
 
 def replay_max_messages_for_context(context_window_tokens: int | None) -> int:
     if not context_window_tokens or context_window_tokens <= 0:
-        return DEFAULT_REPLAY_MAX_MESSAGES
+        return FILE_MAX_MESSAGES
     return min(
         FILE_MAX_MESSAGES,
         max(MIN_REPLAY_MAX_MESSAGES, context_window_tokens // REPLAY_TOKENS_PER_MESSAGE),
@@ -144,7 +143,7 @@ class Session:
 
     def get_history(
         self,
-        max_messages: int = DEFAULT_REPLAY_MAX_MESSAGES,
+        max_messages: int = FILE_MAX_MESSAGES,
         *,
         max_tokens: int = 0,
         extend_to_user: bool = False,
@@ -155,7 +154,7 @@ class Session:
         token budget from the tail (``max_tokens``) when provided.
         """
         unconsolidated = self.messages[self.last_consolidated:]
-        max_messages = max_messages if max_messages > 0 else DEFAULT_REPLAY_MAX_MESSAGES
+        max_messages = max_messages if max_messages > 0 else FILE_MAX_MESSAGES
         start_idx = recent_message_start_index(
             unconsolidated,
             max_messages,