refactor(backend): Update references from document summary LLM to agent LLM across multiple files

surfsense_backend/app/agents/multi_agent_chat/subagents/builtins/deliverables/tools/report.py

@@ -546,7 +546,7 @@ def create_generate_report_tool(
     Factory function to create the generate_report tool with injected dependencies.
 
     The tool generates a Markdown report inline using the search space's
-    document summary LLM, saves it to the database, and returns immediately.
+    agent LLM, saves it to the database, and returns immediately.
 
     Uses short-lived database sessions for each DB operation so no connection
     is held during the long LLM API call.

surfsense_backend/app/agents/new_chat/tools/report.py

@@ -565,7 +565,7 @@ def create_generate_report_tool(
     Factory function to create the generate_report tool with injected dependencies.
 
     The tool generates a Markdown report inline using the search space's
-    document summary LLM, saves it to the database, and returns immediately.
+    agent LLM, saves it to the database, and returns immediately.
 
     Uses short-lived database sessions for each DB operation so no connection
     is held during the long LLM API call.

surfsense_backend/app/agents/podcaster/nodes.py

@@ -31,12 +31,10 @@ async def create_podcast_transcript(
     search_space_id = configuration.search_space_id
     user_prompt = configuration.user_prompt
 
-    # Get search space's document summary LLM
+    # Use the search space's agent LLM for podcast transcript generation.
     llm = await get_agent_llm(state.db_session, search_space_id)
     if not llm:
-        error_message = (
+        error_message = f"No agent LLM configured for search space {search_space_id}"
-            f"No document summary LLM configured for search space {search_space_id}"
-        )
         print(error_message)
         raise RuntimeError(error_message)
 

surfsense_backend/app/celery_app.py

@@ -103,7 +103,7 @@ def init_worker(**kwargs):
     """Initialize the LLM Router and Image Gen Router when a Celery worker process starts.
 
     This ensures the Auto mode (LiteLLM Router) is available for background tasks
-    like document summarization and image generation.
+    like agent workflows and image generation.
     """
     from app.observability.bootstrap import init_otel
 

surfsense_backend/app/db.py

@@ -2963,7 +2963,7 @@ async def shielded_async_session():
 async def setup_indexes():
     async with engine.begin() as conn:
         # Create indexes
-        # Document Summary Indexes
+        # Document embedding indexes
         await conn.execute(
             text(
                 "CREATE INDEX IF NOT EXISTS document_vector_index ON documents USING hnsw (embedding public.vector_cosine_ops)"

surfsense_backend/app/tasks/connector_indexers/google_calendar_indexer.py

@@ -2,7 +2,7 @@
 Google Calendar connector indexer.
 
 Uses the shared IndexingPipelineService for document deduplication,
-summarization, chunking, and embedding.
+chunking, and embedding.
 """
 
 from collections.abc import Awaitable, Callable

surfsense_backend/app/tasks/connector_indexers/google_gmail_indexer.py

@@ -2,7 +2,7 @@
 Google Gmail connector indexer.
 
 Uses the shared IndexingPipelineService for document deduplication,
-summarization, chunking, and embedding.
+chunking, and embedding.
 """
 
 from collections.abc import Awaitable, Callable

surfsense_backend/app/tasks/connector_indexers/linear_indexer.py

@@ -2,7 +2,7 @@
 Linear connector indexer.
 
 Uses the shared IndexingPipelineService for document deduplication,
-summarization, chunking, and embedding with bounded parallel indexing.
+chunking, and embedding with bounded parallel indexing.
 """
 
 from collections.abc import Awaitable, Callable

surfsense_backend/app/tasks/connector_indexers/notion_indexer.py

@@ -2,7 +2,7 @@
 Notion connector indexer.
 
 Uses the shared IndexingPipelineService for document deduplication,
-summarization, chunking, and embedding with bounded parallel indexing.
+chunking, and embedding with bounded parallel indexing.
 """
 
 from collections.abc import Awaitable, Callable