Merge main into feature/support-single-agent-transcription

Author: jinnigu

.github/pull_request_template.md

@@ -0,0 +1,52 @@
+**Please ensure you have read the [contribution guide](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) before creating a pull request.**
+
+### Link to Issue or Description of Change
+
+**1. Link to an existing issue (if applicable):**
+
+- Closes: #_issue_number_
+- Related: #_issue_number_
+
+**2. Or, if no issue exists, describe the change:**
+
+_If applicable, please follow the issue templates to provide as much detail as
+possible._
+
+**Problem:**
+_A clear and concise description of what the problem is._
+
+**Solution:**
+_A clear and concise description of what you want to happen and why you choose
+this solution._
+
+### Testing Plan
+
+_Please describe the tests that you ran to verify your changes. This is required
+for all PRs that are not small documentation or typo fixes._
+
+**Unit Tests:**
+
+- [ ] I have added or updated unit tests for my change.
+- [ ] All unit tests pass locally.
+
+_Please include a summary of passed `pytest` results._
+
+**Manual End-to-End (E2E) Tests:**
+
+_Please provide instructions on how to manually test your changes, including any
+necessary setup or configuration. Please provide logs or screenshots to help
+reviewers better understand the fix._
+
+### Checklist
+
+- [ ] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document.
+- [ ] I have performed a self-review of my own code.
+- [ ] I have commented my code, particularly in hard-to-understand areas.
+- [ ] I have added tests that prove my fix is effective or that my feature works.
+- [ ] New and existing unit tests pass locally with my changes.
+- [ ] I have manually tested my changes end-to-end.
+- [ ] Any dependent changes have been merged and published in downstream modules.
+
+### Additional context
+
+_Add any other context or screenshots about the feature request here._

.github/workflows/pr-triage.yml

@@ -3,10 +3,16 @@ name: ADK Pull Request Triaging Agent
 on:
   pull_request_target:
     types: [opened, reopened, edited]
+  workflow_dispatch:
+    inputs:
+      pr_number:
+        description: 'The Pull Request number to triage'
+        required: true
+        type: 'string'
 
 jobs:
   agent-triage-pull-request:
-    if: "!contains(github.event.pull_request.labels.*.name, 'bot triaged') && !contains(github.event.pull_request.labels.*.name, 'google-contributor')"
+    if: github.event_name == 'workflow_dispatch' || !contains(github.event.pull_request.labels.*.name, 'google-contributor')
     runs-on: ubuntu-latest
     permissions:
       pull-requests: write
@@ -33,7 +39,7 @@ jobs:
           GOOGLE_GENAI_USE_VERTEXAI: 0
           OWNER: 'google'
           REPO: 'adk-python'
-          PULL_REQUEST_NUMBER: ${{ github.event.pull_request.number }}
+          PULL_REQUEST_NUMBER: ${{ github.event.pull_request.number || github.event.inputs.pr_number }}
           INTERACTIVE: ${{ vars.PR_TRIAGE_INTERACTIVE }}
           PYTHONPATH: contributing/samples
         run: python -m adk_pr_triaging_agent.main

.github/workflows/triage.yml

@@ -3,8 +3,6 @@ name: ADK Issue Triaging Agent
 on:
   issues:
     types: [opened, reopened]
-  schedule:
-    - cron: '0 */6 * * *' # every 6h
 
 jobs:
   agent-triage-issues:

CHANGELOG.md

@@ -57,7 +57,9 @@ information on using pull requests.
 ### Requirement for PRs
 
 -   All PRs, other than small documentation or typo fixes, should have a Issue
-    associated. If not, please create one.
+    associated. If a relevant issue doesn't exist, please create one first or
+    you may instead describe the bug or feature directly within the PR
+    description, following the structure of our issue templates.
 -   Small, focused PRs. Keep changes minimal—one concern per PR.
 -   For bug fixes or features, please provide logs or screenshot after the fix
     is applied to help reviewers better understand the fix.

CONTRIBUTING.md

@@ -24,15 +24,15 @@
 
 Agent Development Kit (ADK) is a flexible and modular framework for developing and deploying AI agents. While optimized for Gemini and the Google ecosystem, ADK is model-agnostic, deployment-agnostic, and is built for compatibility with other frameworks. ADK was designed to make agent development feel more like software development, to make it easier for developers to create, deploy, and orchestrate agentic architectures that range from simple tasks to complex workflows.
 
-
 ---
 
 ## 🔥 What's new
 
-- **Agent Config**: Build agents without code. Check out the
-  [Agent Config](https://google.github.io/adk-docs/agents/config/) feature.
+- **Custom Service Registration**: Add a service registry to provide a generic way to register custom service implementations to be used in FastAPI server. See short instruction [here](https://github.com/google/adk-python/discussions/3175#discussioncomment-14745120). ([391628f](https://github.com/google/adk-python/commit/391628fcdc7b950c6835f64ae3ccab197163c990))
+
+- **Rewind**: Add the ability to rewind a session to before a previous invocation ([9dce06f](https://github.com/google/adk-python/commit/9dce06f9b00259ec42241df4f6638955e783a9d1)).
 
-- **Tool Confirmation**: A [tool confirmation flow(HITL)](https://google.github.io/adk-docs/tools/confirmation/) that can guard tool execution with explicit confirmation and custom input
+- **New CodeExecutor**: Introduces a new AgentEngineSandboxCodeExecutor class that supports executing agent-generated code using the Vertex AI Code Execution Sandbox API ([ee39a89](https://github.com/google/adk-python/commit/ee39a891106316b790621795b5cc529e89815a98))
 
 ## ✨ Key Features
 
@@ -43,6 +43,11 @@ Agent Development Kit (ADK) is a flexible and modular framework for developing a
 - **Code-First Development**: Define agent logic, tools, and orchestration
   directly in Python for ultimate flexibility, testability, and versioning.
 
+- **Agent Config**: Build agents without code. Check out the
+  [Agent Config](https://google.github.io/adk-docs/agents/config/) feature.
+
+- **Tool Confirmation**: A [tool confirmation flow(HITL)](https://google.github.io/adk-docs/tools/confirmation/) that can guard tool execution with explicit confirmation and custom input.
+
 - **Modular Multi-Agent Systems**: Design scalable applications by composing
   multiple specialized agents into flexible hierarchies.
 

README.md

@@ -24,15 +24,16 @@
 from google.adk.models import BaseLlm
 from google.adk.tools import AgentTool
 from google.adk.tools import FunctionTool
+from google.genai import types
 
+from .sub_agents.adk_knowledge_agent import create_adk_knowledge_agent
 from .sub_agents.google_search_agent import create_google_search_agent
 from .sub_agents.url_context_agent import create_url_context_agent
 from .tools.cleanup_unused_files import cleanup_unused_files
 from .tools.delete_files import delete_files
 from .tools.explore_project import explore_project
 from .tools.read_config_files import read_config_files
 from .tools.read_files import read_files
-from .tools.resolve_root_directory import resolve_root_directory
 from .tools.search_adk_source import search_adk_source
 from .tools.write_config_files import write_config_files
 from .tools.write_files import write_files
@@ -58,9 +59,7 @@ def create_agent(
       Configured LlmAgent with embedded ADK AgentConfig schema
     """
     # Load full ADK AgentConfig schema directly into instruction context
-    instruction = AgentBuilderAssistant._load_instruction_with_schema(
-        model, working_directory
-    )
+    instruction = AgentBuilderAssistant._load_instruction_with_schema(model)
 
     # TOOL ARCHITECTURE: Hybrid approach using both AgentTools and FunctionTools
     #
@@ -70,9 +69,14 @@ def create_agent(
     # - Maintains compatibility with existing ADK tool ecosystem
 
     # Built-in ADK tools wrapped as sub-agents
+    adk_knowledge_agent = create_adk_knowledge_agent()
     google_search_agent = create_google_search_agent()
     url_context_agent = create_url_context_agent()
-    agent_tools = [AgentTool(google_search_agent), AgentTool(url_context_agent)]
+    agent_tools = [
+        AgentTool(adk_knowledge_agent),
+        AgentTool(google_search_agent),
+        AgentTool(url_context_agent),
+    ]
 
     # CUSTOM FUNCTION TOOLS: Agent Builder specific capabilities
     #
@@ -88,8 +92,6 @@ def create_agent(
             write_config_files
         ),  # Write/validate multiple YAML configs
         FunctionTool(explore_project),  # Analyze project structure
-        # Working directory context tools
-        FunctionTool(resolve_root_directory),
         # File management tools (multi-file support)
         FunctionTool(read_files),  # Read multiple files
         FunctionTool(write_files),  # Write multiple files
@@ -112,6 +114,9 @@ def create_agent(
         instruction=instruction,
         model=model,
         tools=all_tools,
+        generate_content_config=types.GenerateContentConfig(
+            max_output_tokens=8192,
+        ),
     )
 
     return agent
@@ -125,7 +130,6 @@ def _load_schema() -> str:
     # ADK AgentConfig schema loading with caching and error handling.
     schema_content = load_agent_config_schema(
         raw_format=True,  # Get as JSON string
-        escape_braces=True,  # Escape braces for template embedding
     )
 
     # Format as indented code block for instruction embedding
@@ -147,7 +151,6 @@ def _load_schema() -> str:
   @staticmethod
   def _load_instruction_with_schema(
       model: Union[str, BaseLlm],
-      working_directory: Optional[str] = None,
   ) -> Callable[[ReadonlyContext], str]:
     """Load instruction template and embed ADK AgentConfig schema content."""
     instruction_template = (
@@ -162,19 +165,43 @@ def _load_instruction_with_schema(
         else getattr(model, "model_name", str(model))
     )
 
-    # Fill the instruction template with ADK AgentConfig schema content and default model
-    instruction_text = instruction_template.format(
-        schema_content=schema_content, default_model=model_str
-    )
-
     # Return a function that accepts ReadonlyContext and returns the instruction
     def instruction_provider(context: ReadonlyContext) -> str:
-      return AgentBuilderAssistant._compile_instruction_with_context(
-          instruction_text, context, working_directory
+      # Extract project folder name from session state
+      project_folder_name = AgentBuilderAssistant._extract_project_folder_name(
+          context
+      )
+
+      # Fill the instruction template with all variables
+      instruction_text = instruction_template.format(
+          schema_content=schema_content,
+          default_model=model_str,
+          project_folder_name=project_folder_name,
       )
+      return instruction_text
 
     return instruction_provider
 
+  @staticmethod
+  def _extract_project_folder_name(context: ReadonlyContext) -> str:
+    """Extract project folder name from session state using resolve_file_path."""
+    from .utils.resolve_root_directory import resolve_file_path
+
+    session_state = context._invocation_context.session.state
+
+    # Use resolve_file_path to get the full resolved path for "."
+    # This handles all the root_directory resolution logic consistently
+    resolved_path = resolve_file_path(".", session_state)
+
+    # Extract the project folder name from the resolved path
+    project_folder_name = resolved_path.name
+
+    # Fallback to "project" if we somehow get an empty name
+    if not project_folder_name:
+      project_folder_name = "project"
+
+    return project_folder_name
+
   @staticmethod
   def _load_embedded_schema_instruction_template() -> str:
     """Load instruction template for embedded ADK AgentConfig schema mode."""
@@ -187,70 +214,3 @@ def _load_embedded_schema_instruction_template() -> str:
 
     with open(template_path, "r", encoding="utf-8") as f:
       return f.read()
-
-  @staticmethod
-  def _compile_instruction_with_context(
-      instruction_text: str,
-      context: ReadonlyContext,
-      working_directory: Optional[str] = None,
-  ) -> str:
-    """Compile instruction with session context and working directory information.
-
-    This method enhances instructions with:
-    1. Working directory information for path resolution
-    2. Session-based root directory binding if available
-
-    Args:
-      instruction_text: Base instruction text
-      context: ReadonlyContext from the agent session
-      working_directory: Optional working directory for path resolution
-
-    Returns:
-      Enhanced instruction text with context information
-    """
-    import os
-
-    # Get working directory (use provided or current working directory)
-    actual_working_dir = working_directory or os.getcwd()
-
-    # Check for existing root directory in session state
-    session_root_directory = context._invocation_context.session.state.get(
-        "root_directory"
-    )
-
-    # Compile additional context information
-    context_info = f"""
-
-## SESSION CONTEXT
-
-**Working Directory**: `{actual_working_dir}`
-- Use this as the base directory for path resolution when calling resolve_root_directory
-- Pass this as the working_directory parameter to resolve_root_directory tool
-
-"""
-
-    if session_root_directory:
-      context_info += f"""**Established Root Directory**: `{session_root_directory}`
-- This session is bound to root directory: {session_root_directory}
-- DO NOT ask the user for root directory - use this established path
-- All agent building should happen within this root directory
-- If user wants to work in a different directory, ask them to start a new chat session
-
-"""
-    else:
-      context_info += f"""**Root Directory**: Not yet established
-- You MUST ask the user for their desired root directory first
-- Use resolve_root_directory tool to validate the path
-- Once confirmed, this session will be bound to that root directory
-
-"""
-
-    context_info += """**Session Binding Rules**:
-- Each chat session is bound to ONE root directory
-- Once established, work only within that root directory
-- To switch directories, user must start a new chat session
-- Always verify paths using resolve_root_directory tool before creating files
-
-"""
-
-    return instruction_text + context_info