[pull] main from google:main

.github/workflows/analyze-releases-for-adk-docs-updates.yml

@@ -16,7 +16,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python
         uses: actions/setup-python@v6

.github/workflows/check-file-contents.yml

@@ -24,7 +24,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout Code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
         with:
           fetch-depth: 2
 

.github/workflows/discussion_answering.yml

@@ -15,7 +15,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python
         uses: actions/setup-python@v6
@@ -24,7 +24,7 @@ jobs:
 
       - name: Authenticate to Google Cloud
         id: auth
-        uses: 'google-github-actions/auth@v2'
+        uses: 'google-github-actions/auth@v3'
         with:
           credentials_json: '${{ secrets.ADK_GCP_SA_KEY }}'
 

.github/workflows/isort.yml

@@ -26,7 +26,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
         with:
           fetch-depth: 2
 

.github/workflows/pr-triage.yml

@@ -20,7 +20,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python
         uses: actions/setup-python@v6

.github/workflows/pyink.yml

@@ -26,7 +26,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
         with:
           fetch-depth: 2
 

.github/workflows/python-unit-tests.yml

@@ -29,7 +29,7 @@ jobs:
 
     steps:
     - name: Checkout code
-      uses: actions/checkout@v5
+      uses: actions/checkout@v6
 
     - name: Set up Python ${{ matrix.python-version }}
       uses: actions/setup-python@v6

.github/workflows/stale-bot.yml

@@ -18,7 +18,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python
         uses: actions/setup-python@v6

.github/workflows/triage.yml

@@ -24,7 +24,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Set up Python
         uses: actions/setup-python@v6

.github/workflows/upload-adk-docs-to-vertex-ai-search.yml

@@ -13,7 +13,7 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Clone adk-docs repository
         run: git clone https://github.com/google/adk-docs.git /tmp/adk-docs
@@ -28,7 +28,7 @@ jobs:
 
       - name: Authenticate to Google Cloud
         id: auth
-        uses: 'google-github-actions/auth@v2'
+        uses: 'google-github-actions/auth@v3'
         with:
           credentials_json: '${{ secrets.ADK_GCP_SA_KEY }}'
 

contributing/samples/api_registry_agent/agent.py

@@ -32,8 +32,19 @@
 root_agent = LlmAgent(
     model="gemini-2.0-flash",
     name="bigquery_assistant",
-    instruction="""
-Help user access their BigQuery data via API Registry tools.
+    instruction=f"""
+You are a helpful data analyst assistant with access to BigQuery. The project ID is: {PROJECT_ID}
+
+When users ask about data:
+- Use the project ID {PROJECT_ID} when calling BigQuery tools.
+- First, explore available datasets and tables to understand what data exists.
+- Check table schemas to understand the structure before querying.
+- Write clear, efficient SQL queries to answer their questions.
+- Explain your findings in simple, non-technical language.
+
+Mandatory Requirements:
+- Always use the BigQuery tools to fetch real data rather than making assumptions.
+- For all BigQuery operations, use project_id: {PROJECT_ID}.
     """,
     tools=[registry_tools],
 )

src/google/adk/cli/utils/local_storage.py

@@ -34,6 +34,8 @@
 
 logger = logging.getLogger("google_adk." + __name__)
 
+_BUILT_IN_SESSION_SERVICE_KEY = "__adk_built_in_session_service__"
+
 
 def create_local_database_session_service(
     *,
@@ -124,6 +126,16 @@ def __init__(
 
   async def _get_service(self, app_name: str) -> BaseSessionService:
     async with self._service_lock:
+      if app_name.startswith("__"):
+        service = self._services.get(_BUILT_IN_SESSION_SERVICE_KEY)
+        if service is not None:
+          return service
+        service = create_local_database_session_service(
+            base_dir=self._agents_root,
+        )
+        self._services[_BUILT_IN_SESSION_SERVICE_KEY] = service
+        return service
+
       storage_name = self._app_name_to_dir.get(app_name, app_name)
       service = self._services.get(storage_name)
       if service is not None:

src/google/adk/models/lite_llm.py

@@ -628,9 +628,12 @@ def _is_ollama_chat_provider(
     model: Optional[str], custom_llm_provider: Optional[str]
 ) -> bool:
   """Returns True when requests should be normalized for ollama_chat."""
-  if custom_llm_provider and custom_llm_provider.lower() == "ollama_chat":
+  if (
+      custom_llm_provider
+      and custom_llm_provider.strip().lower() == "ollama_chat"
+  ):
     return True
-  if model and model.lower().startswith("ollama_chat"):
+  if model and model.strip().lower().startswith("ollama_chat"):
     return True
   return False
 
@@ -644,11 +647,24 @@ def _flatten_ollama_content(
   join them with newlines, and fall back to a JSON string for non-text content.
   If both text and non-text parts are present, only the text parts are kept.
   """
-  if not isinstance(content, list):
+  if content is None or isinstance(content, str):
     return content
 
+  # `OpenAIMessageContent` is typed as `Iterable[...]` in LiteLLM. Some
+  # providers or LiteLLM versions may hand back tuples or other iterables.
+  if isinstance(content, dict):
+    try:
+      return json.dumps(content)
+    except TypeError:
+      return str(content)
+
+  try:
+    blocks = list(content)
+  except TypeError:
+    return str(content)
+
   text_parts = []
-  for block in content:
+  for block in blocks:
     if isinstance(block, dict) and block.get("type") == "text":
       text_value = block.get("text")
       if text_value:
@@ -658,9 +674,9 @@ def _flatten_ollama_content(
     return _NEW_LINE.join(text_parts)
 
   try:
-    return json.dumps(content)
+    return json.dumps(blocks)
   except TypeError:
-    return str(content)
+    return str(blocks)
 
 
 def _normalize_ollama_chat_messages(

tests/unittests/cli/utils/test_local_storage.py

@@ -72,6 +72,18 @@ async def test_per_agent_session_service_respects_app_name_alias(
   assert (tmp_path / folder_name / ".adk" / "session.db").exists()
 
 
+@pytest.mark.asyncio
+async def test_per_agent_session_service_routes_built_in_agents_to_root_dot_adk(
+    tmp_path: Path,
+) -> None:
+  service = PerAgentDatabaseSessionService(agents_root=tmp_path)
+
+  await service.create_session(app_name="__helper", user_id="user")
+
+  assert not (tmp_path / "__helper").exists()
+  assert (tmp_path / ".adk" / "session.db").exists()
+
+
 def test_create_local_database_session_service_returns_sqlite(
     tmp_path: Path,
 ) -> None:

tests/unittests/models/test_litellm.py

@@ -1549,6 +1549,17 @@ async def test_generate_content_async_custom_provider_flattens_content(
   assert "Describe this image." in message_content
 
 
+def test_flatten_ollama_content_accepts_tuple_blocks():
+  from google.adk.models.lite_llm import _flatten_ollama_content
+
+  content = (
+      {"type": "text", "text": "first"},
+      {"type": "text", "text": "second"},
+  )
+  flattened = _flatten_ollama_content(content)
+  assert flattened == "first\nsecond"
+
+
 @pytest.mark.asyncio
 async def test_content_to_message_param_user_message():
   content = types.Content(