fix: include pending summaries in consolidation estimates

Made-with: Cursor

nanobot/agent/loop.py

@@ -648,7 +648,10 @@ class AgentLoop:
 
             session, pending = self.auto_compact.prepare_session(session, key)
 
-            await self.consolidator.maybe_consolidate_by_tokens(session)
+            await self.consolidator.maybe_consolidate_by_tokens(
+                session,
+                session_summary=pending,
+            )
             # Persist subagent follow-ups into durable history BEFORE prompt
             # assembly. ContextBuilder merges adjacent same-role messages for
             # provider compatibility, which previously caused the follow-up to
@@ -709,7 +712,10 @@ class AgentLoop:
         if result := await self.commands.dispatch(ctx):
             return result
 
-        await self.consolidator.maybe_consolidate_by_tokens(session)
+        await self.consolidator.maybe_consolidate_by_tokens(
+            session,
+            session_summary=pending,
+        )
 
         self._set_tool_context(msg.channel, msg.chat_id, msg.metadata.get("message_id"))
         if message_tool := self.tools.get("message"):

nanobot/agent/memory.py

@@ -416,7 +416,12 @@ class Consolidator:
                 return idx
         return None
 
-    def estimate_session_prompt_tokens(self, session: Session) -> tuple[int, str]:
+    def estimate_session_prompt_tokens(
+        self,
+        session: Session,
+        *,
+        session_summary: str | None = None,
+    ) -> tuple[int, str]:
         """Estimate current prompt size for the normal session history view."""
         history = session.get_history(max_messages=0)
         channel, chat_id = (session.key.split(":", 1) if ":" in session.key else (None, None))
@@ -425,6 +430,7 @@ class Consolidator:
             current_message="[token-probe]",
             channel=channel,
             chat_id=chat_id,
+            session_summary=session_summary,
         )
         return estimate_prompt_tokens_chain(
             self.provider,
@@ -467,7 +473,12 @@ class Consolidator:
             self.store.raw_archive(messages)
             return None
 
-    async def maybe_consolidate_by_tokens(self, session: Session) -> None:
+    async def maybe_consolidate_by_tokens(
+        self,
+        session: Session,
+        *,
+        session_summary: str | None = None,
+    ) -> None:
         """Loop: archive old messages until prompt fits within safe budget.
 
         The budget reserves space for completion tokens and a safety buffer
@@ -481,7 +492,10 @@ class Consolidator:
             budget = self.context_window_tokens - self.max_completion_tokens - self._SAFETY_BUFFER
             target = budget // 2
             try:
-                estimated, source = self.estimate_session_prompt_tokens(session)
+                estimated, source = self.estimate_session_prompt_tokens(
+                    session,
+                    session_summary=session_summary,
+                )
             except Exception:
                 logger.exception("Token estimation failed for {}", session.key)
                 estimated, source = 0, "error"
@@ -545,7 +559,10 @@ class Consolidator:
                 self.sessions.save(session)
 
                 try:
-                    estimated, source = self.estimate_session_prompt_tokens(session)
+                    estimated, source = self.estimate_session_prompt_tokens(
+                        session,
+                        session_summary=session_summary,
+                    )
                 except Exception:
                     logger.exception("Token estimation failed for {}", session.key)
                     estimated, source = 0, "error"