feat(agents): make sources optional and drop the synthetic "Default" source

The sources list used to start with a fake "Default" entry that had no id
and, at run time, meant "no source, skip retrieval". The agent form
pre-selected it, snapped back to it when the last source was deselected,
and refused to publish without it, so a new agent always looked like it
had a knowledge base when it had none.

Backend
- /api/sources returns only ingested sources; no placeholder row.
- Publishing an agent no longer requires a source on create or update.
  The legacy "default" value is still accepted and maps to NULL.

Frontend
- The agent source picker starts empty, can be cleared, and shows a hint
  that a source-less agent answers from the model and its tools only.
- The picker groups sources into "Your sources" and "Shared with team"
  when any team-shared source exists, shows "N sources selected" for a
  multi-selection, and gets the same "Go to Sources" / "Upload new"
  footer as the chat picker. A source uploaded from the form is selected
  when it lands.
- Source selection serialisation and the picker id live in one helper
  shared with the chat picker; the four copies in the form are gone.
- The client no longer seeds a placeholder source in the store, and the
  dead auto-select of a "default" document is removed.

docsgpt/api/answer/services/stream_processor.py

@@ -647,9 +647,9 @@ class StreamProcessor:
     def _configure_source(self):
         """Configure the source based on agent data.
 
-        The literal string ``"default"`` is a placeholder meaning "no
-        ingested source" and is normalized to an empty source so that no
-        retrieval is attempted.
+        The literal string ``"default"`` is a legacy placeholder (older
+        clients sent it for "no ingested source") and is normalized to an
+        empty source so that no retrieval is attempted.
         """
         if self._agent_data:
             agent_data = self._agent_data

docsgpt/api/user/agents/routes.py

@@ -76,7 +76,8 @@ AGENT_TYPE_SCHEMAS = {
         # is acceptable and maps to NULL downstream.
         "validate_published": ["name", "description"],
         "validate_draft": [],
-        "require_source": True,
+        # No source is required: a source-less agent answers from the model
+        # and its tools, and ``source``/``sources`` map to NULL downstream.
         "fields": [
             "name",
             "description",
@@ -661,20 +662,6 @@ class CreateAgent(Resource):
         if data.get("status") == "published":
             required_fields = schema["required_published"]
             validate_fields = schema["validate_published"]
-            if (
-                schema.get("require_source")
-                and not data.get("source")
-                and not data.get("sources")
-            ):
-                return make_response(
-                    jsonify(
-                        {
-                            "success": False,
-                            "message": "Either 'source' or 'sources' field is required for published agents",
-                        }
-                    ),
-                    400,
-                )
         else:
             required_fields = schema["required_draft"]
             validate_fields = schema["validate_draft"]
@@ -1190,29 +1177,9 @@ class UpdateAgent(Resource):
                             )
                             if not final_value:
                                 missing_published_fields.append(field_label)
-                        source_final = update_fields.get(
-                            "source_id", existing_agent.get("source_id"),
-                        )
-                        extra_final = update_fields.get(
-                            "extra_source_ids", existing_agent.get("extra_source_ids") or [],
-                        )
-                        # ``retriever`` carries the runtime identity for
-                        # agents that publish against the synthetic
-                        # "Default" source (frontend's auto-selected
-                        # ``{name: "Default", retriever: "classic"}``
-                        # entry has no ``id``, so ``source_id`` ends up
-                        # NULL even though the user picked something).
-                        # Without this fallback the most common new-agent
-                        # publish flow gets a 400.
-                        retriever_final = update_fields.get(
-                            "retriever", existing_agent.get("retriever"),
-                        )
-                        if (
-                            not source_final
-                            and not extra_final
-                            and not retriever_final
-                        ):
-                            missing_published_fields.append("Source or retriever")
+                        # Sources are optional: a published agent with no
+                        # ``source_id`` and no ``extra_source_ids`` skips
+                        # retrieval and answers from the model and its tools.
                         if missing_published_fields:
                             return _reject(
                                 "Cannot publish agent. Missing or invalid required "

docsgpt/api/user/sources/routes.py

@@ -73,16 +73,10 @@ class CombinedJson(Resource):
         if not decoded_token:
             return make_response(jsonify({"success": False}), 401)
         user = decoded_token.get("sub")
-        data = [
-            {
-                "name": "Default",
-                "date": "default",
-                "model": settings.EMBEDDINGS_NAME,
-                "location": "remote",
-                "tokens": "",
-                "retriever": "classic",
-            }
-        ]
+        # Only ingested sources are listed. There is no synthetic "Default"
+        # entry: "no source" is expressed by an empty selection on the client,
+        # and the answer path treats a missing source as "skip retrieval".
+        data = []
 
         try:
             with db_readonly() as conn: