chore: Bump version and update CHANGELOG for v1.9.0

PiperOrigin-RevId: 789494584

.github/workflows/pr-commit-check.yml

@@ -1,62 +0,0 @@
-# .github/workflows/pr-commit-check.yml
-# This GitHub Action workflow checks if a pull request has more than one commit.
-# If it does, it fails the check and instructs the user to squash their commits.
-
-name: 'PR Commit Check'
-
-# This workflow runs on pull request events.
-# It's configured to run on any pull request that is opened or synchronized (new commits pushed).
-on:
-  pull_request:
-    types: [opened, synchronize]
-
-# Defines the jobs that will run as part of the workflow.
-jobs:
-  check-commit-count:
-    # The type of runner that the job will run on. 'ubuntu-latest' is a good default.
-    runs-on: ubuntu-latest
-
-    # The steps that will be executed as part of the job.
-    steps:
-      # Step 1: Check out the code
-      # This action checks out your repository under $GITHUB_WORKSPACE, so your workflow can access it.
-      - name: Checkout Code
-        uses: actions/checkout@v4
-        with:
-          # We need to fetch all commits to accurately count them.
-          # '0' means fetch all history for all branches and tags.
-          fetch-depth: 0
-
-      # Step 2: Count the commits in the pull request
-      # This step runs a script to get the number of commits in the PR.
-      - name: Count Commits
-        id: count_commits
-        # We use `git rev-list --count` to count the commits.
-        # ${{ github.event.pull_request.base.sha }} is the commit SHA of the base branch.
-        # ${{ github.event.pull_request.head.sha }} is the commit SHA of the head branch (the PR branch).
-        # The '..' syntax gives us the list of commits in the head branch that are not in the base branch.
-        # The output of the command (the count) is stored in a step output variable named 'count'.
-        run: |
-          count=$(git rev-list --count ${{ github.event.pull_request.base.sha }}..${{ github.event.pull_request.head.sha }})
-          echo "commit_count=$count" >> $GITHUB_OUTPUT
-
-      # Step 3: Check if the commit count is greater than 1
-      # This step uses the output from the previous step to decide whether to pass or fail.
-      - name: Check Commit Count
-        # This step only runs if the 'commit_count' output from the 'count_commits' step is greater than 1.
-        if: steps.count_commits.outputs.commit_count > 1
-        # If the condition is met, the workflow will exit with a failure status.
-        run: |
-          echo "This pull request has ${{ steps.count_commits.outputs.commit_count }} commits."
-          echo "Please squash them into a single commit before merging."
-          echo "You can use git rebase -i HEAD~N"
-          echo "...where N is the number of commits you want to squash together. The PR check conveniently tells you this number! For example, if the check says you have 3 commits, you would run: git rebase -i HEAD~3."
-          echo "Because you have rewritten the commit history, you must use the --force flag to update the pull request: git push --force"
-          exit 1
-
-      # Step 4: Success message
-      # This step runs if the commit count is not greater than 1 (i.e., it's 1).
-      - name: Success
-        if: steps.count_commits.outputs.commit_count <= 1
-        run: |
-          echo "This pull request has a single commit. Great job!"

.github/workflows/pr-triage.yml

@@ -0,0 +1,38 @@
+name: ADK Pull Request Triaging Agent
+
+on:
+  pull_request_target:
+    types: [opened, reopened, edited]
+
+jobs:
+  agent-triage-pull-request:
+    runs-on: ubuntu-latest
+    permissions:
+      pull-requests: write
+      contents: read
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.11'
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install requests google-adk
+
+      - name: Run Triaging Script
+        env:
+          GITHUB_TOKEN: ${{ secrets.ADK_TRIAGE_AGENT }}
+          GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }}
+          GOOGLE_GENAI_USE_VERTEXAI: 0
+          OWNER: 'google'
+          REPO: 'adk-python'
+          PULL_REQUEST_NUMBER: ${{ github.event.pull_request.number }}
+          INTERACTIVE: ${{ vars.PR_TRIAGE_INTERACTIVE }}
+          PYTHONPATH: contributing/samples
+        run: python -m adk_pr_triaging_agent.main

CHANGELOG.md

@@ -1,5 +1,83 @@
 # Changelog
 
+## [1.9.0](https://github.com/google/adk-python/compare/v1.8.0...v1.9.0) (2025-07-31)
+
+
+### Features
+
+* [CLI] Add `-v`, `--verbose` flag to enable DEBUG logging as a shortcut for `--log_level DEBUG` ([3be0882](https://github.com/google/adk-python/commit/3be0882c63bf9b185c34bcd17e03769b39f0e1c5))
+* [CLI] Add a CLI option to update an agent engine instance ([206a132](https://github.com/google/adk-python/commit/206a13271e5f1bb0bb8114b3bb82f6ec3f030cd7))
+* [CLI] Modularize fast_api.py to allow simpler construction of API Server ([bfc203a](https://github.com/google/adk-python/commit/bfc203a92fdfbc4abaf776e76dca50e7ca59127b), [dfc25c1](https://github.com/google/adk-python/commit/dfc25c17a98aaad81e1e2f140db83d17cd78f393), [e176f03](https://github.com/google/adk-python/commit/e176f03e8fe13049187abd0f14e63afca9ccff01))
+* [CLI] Refactor AgentLoader into base class and add InMemory impl alongside existing filesystem impl ([bda3df2](https://github.com/google/adk-python/commit/bda3df24802d0456711a5cd05544aea54a13398d))
+* [CLI] Respect the .ae_ignore file when deploying to agent engine ([f29ab5d](https://github.com/google/adk-python/commit/f29ab5db0563a343d6b8b437a12557c89b7fc98b))
+* [Core] Add new callbacks to handle tool and model errors ([00afaaf](https://github.com/google/adk-python/commit/00afaaf2fc18fba85709754fb1037bb47f647243))
+* [Core] Add sample plugin for logging ([20537e8](https://github.com/google/adk-python/commit/20537e8bfa31220d07662dad731b4432799e1802))
+* [Core] Expose Gemini RetryOptions to client ([1639298](https://github.com/google/adk-python/commit/16392984c51b02999200bd4f1d6781d5ec9054de))
+* [Evals] Added an Fast API new endpoint to serve eval metric info ([c69dcf8](https://github.com/google/adk-python/commit/c69dcf87795c4fa2ad280b804c9b0bd3fa9bf06f))
+* [Evals] Refactored AgentEvaluator and updated it to use LocalEvalService ([1355bd6](https://github.com/google/adk-python/commit/1355bd643ba8f7fd63bcd6a7284cc48e325d138e))
+
+
+### Bug Fixes
+
+* Add absolutize_imports option when deploying to agent engine ([fbe6a7b](https://github.com/google/adk-python/commit/fbe6a7b8d3a431a1d1400702fa534c3180741eb3))
+* Add space to allow adk deploy cloud_run --a2a ([70c4616](https://github.com/google/adk-python/commit/70c461686ec2c60fcbaa384a3f1ea2528646abba))
+* Copy the original function call args before passing it to callback or tools to avoid being modified ([3432b22](https://github.com/google/adk-python/commit/3432b221727b52af2682d5bf3534d533a50325ef))
+* Eval module not found exception string ([7206e0a](https://github.com/google/adk-python/commit/7206e0a0eb546a66d47fb411f3fa813301c56f42))
+* Fix incorrect token count mapping in telemetry ([c8f8b4a](https://github.com/google/adk-python/commit/c8f8b4a20a886a17ce29abd1cfac2858858f907d))
+* Import cli's artifact dependencies directly ([282d67f](https://github.com/google/adk-python/commit/282d67f253935af56fae32428124a385f812c67d))
+* Keep existing header values while merging tracking headers for `llm_request.config.http_options` in `Gemini.generate_content_async` ([6191412](https://github.com/google/adk-python/commit/6191412b07c3b5b5a58cf7714e475f63e89be847))
+* Merge tracking headers even when `llm_request.config.http_options` is not set in `Gemini.generate_content_async` ([ec8dd57](https://github.com/google/adk-python/commit/ec8dd5721aa151cfc033cc3aad4733df002ae9cb))
+* Restore bigquery sample agent to runnable form ([16e8419](https://github.com/google/adk-python/commit/16e8419e32b54298f782ba56827e5139effd8780))
+* Return session state in list_session API endpoint ([314d6a4](https://github.com/google/adk-python/commit/314d6a4f95c6d37c7da3afbc7253570564623322))
+* Runner was expecting Event object instead of Content object when using early exist feature ([bf72426](https://github.com/google/adk-python/commit/bf72426af2bfd5c2e21c410005842e48b773deb3))
+* Unable to acquire impersonated credentials ([9db5d9a](https://github.com/google/adk-python/commit/9db5d9a3e87d363c1bac0f3d8e45e42bd5380d3e))
+* Update `agent_card_builder` to follow grammar rules ([9c0721b](https://github.com/google/adk-python/commit/9c0721beaa526a4437671e6cc70915073be835e3)), closes [#2223](https://github.com/google/adk-python/issues/2223)
+* Use correct type for actions parameter in ApplicationIntegrationToolset ([ce7253f](https://github.com/google/adk-python/commit/ce7253f63ff8e78bccc7805bd84831f08990b881))
+
+
+### Documentation
+
+* Update documents about the information of vibe coding ([0c85587](https://github.com/google/adk-python/commit/0c855877c57775ad5dad930594f9f071164676da))
+
+
+## [1.8.0](https://github.com/google/adk-python/compare/v1.7.0...v1.8.0) (2025-07-23)
+
+### Features
+
+* [Core]Add agent card builder ([18f5bea](https://github.com/google/adk-python/commit/18f5bea411b3b76474ff31bfb2f62742825b45e5))
+* [Core]Add an to_a2a util to convert adk agent to A2A ASGI application ([a77d689](https://github.com/google/adk-python/commit/a77d68964a1c6b7659d6117d57fa59e43399e0c2))
+* [Core]Add camel case converter for agents ([0e173d7](https://github.com/google/adk-python/commit/0e173d736334f8c6c171b3144ac6ee5b7125c846))
+* [Evals]Use LocalEvalService to run all evals in cli and web ([d1f182e](https://github.com/google/adk-python/commit/d1f182e8e68c4a5a4141592f3f6d2ceeada78887))
+* [Evals]Enable FinalResponseMatchV2 metric as an experiment ([36e45cd](https://github.com/google/adk-python/commit/36e45cdab3bbfb653eee3f9ed875b59bcd525ea1))
+* [Models]Add support for `model-optimizer-*` family of models in vertex ([ffe2bdb](https://github.com/google/adk-python/commit/ffe2bdbe4c2ea86cc7924eb36e8e3bb5528c0016))
+* [Services]Added a sample for History Management ([67284fc](https://github.com/google/adk-python/commit/67284fc46667b8c2946762bc9234a8453d48a43c))
+* [Services]Support passing fully qualified agent engine resource name when constructing session service and memory service ([2e77804](https://github.com/google/adk-python/commit/2e778049d0a675e458f4e
+35fe4104ca1298dbfcf))
+* [Tools]Add ComputerUseToolset ([083dcb4](https://github.com/google/adk-python/commit/083dcb44650eb0e6b70219ede731f2fa78ea7d28))
+* [Tools]Allow toolset to process llm_request before tools returned by it ([3643b4a](https://github.com/google/adk-python/commit/3643b4ae196fd9e38e52d5dc9d1cd43ea0733d36))
+* [Tools]Support input/output schema by fully-qualified code reference ([dfee06a](https://github.com/google/adk-python/commit/dfee06ac067ea909251d6fb016f8331065d430e9))
+* [Tools]Enhance LangchainTool to accept more forms of functions ([0ec69d0](https://github.com/google/adk-python/commit/0ec69d05a4016adb72abf9c94f2e9ff4bdd1848c))
+
+### Bug Fixes
+
+* **Attention**: Logging level for some API requests and responses was moved from `INFO` to `DEBUG` ([ff31f57](https://github.com/google/adk-python/commit/ff31f57dc95149f8f309f83f2ec983ef40f1122c))
+  * Please set `--log_level=DEBUG`, if you are interested in having those API request and responses in logs.
+* Add buffer to the write file option ([f2caf2e](https://github.com/google/adk-python/commit/f2caf2eecaf0336495fb42a2166b1b79e57d82d8))
+* Allow current sub-agent to finish execution before exiting the loop agent due to a sub-agent's escalation. ([2aab1cf](https://github.com/google/adk-python/commit/2aab1cf98e1d0e8454764b549fac21475a633409))
+* Check that `mean_score` is a valid float value ([65cb6d6](https://github.com/google/adk-python/commit/65cb6d6bf3278e6c3529938a7b932e3ef6d6c2ae))
+* Handle non-json-serializable values in the `execute_sql` tool ([13ff009](https://github.com/google/adk-python/commit/13ff009d34836a80f107cb43a632df15f7c215e4))
+* Raise `NotFoundError` in `list_eval_sets` function when app_name doesn't exist ([b17d8b6](https://github.com/google/adk-python/commit/b17d8b6e362a5b2a1b6a2dd0cff5e27a71c27925))
+* Fixed serialization of tools with nested schema ([53df35e](https://github.com/google/adk-python/commit/53df35ee58599e9816bd4b9c42ff48457505e599))
+* Set response schema for function tools that returns `None` ([33ac838](https://github.com/google/adk-python/commit/33ac8380adfff46ed8a7d518ae6f27345027c074))
+* Support path level parameters for open_api_spec_parser ([6f01660](https://github.com/google/adk-python/commit/6f016609e889bb0947877f478de0c5729cfcd0c3))
+* Use correct type for actions parameter in ApplicationIntegrationToolset ([ce7253f](https://github.com/google/adk-python/commit/ce7253f63ff8e78bccc7805bd84831f08990b881))
+* Use the same word extractor for query and event contents in InMemoryMemoryService ([1c4c887](https://github.com/google/adk-python/commit/1c4c887bec9326aad2593f016540160d95d03f33))
+
+### Documentation
+
+* Fix missing toolbox-core dependency and improve installation guide ([2486349](https://github.com/google/adk-python/commit/24863492689f36e3c7370be40486555801858bac))
+
+
 ## 1.7.0 (2025-07-16)
 
 ### Features

CONTRIBUTING.md

@@ -210,3 +210,7 @@ All submissions, including submissions by project members, require review. We
 use GitHub pull requests for this purpose. Consult
 [GitHub Help](https://help.github.com/articles/about-pull-requests/) for more
 information on using pull requests.
+
+# Vibe Coding
+
+If you want to contribute by leveraging viber coding, the AGENTS.md (https://github.com/google/adk-python/tree/main/AGENTS.md) could be used as context to your LLM.

README.md

@@ -138,6 +138,10 @@ We welcome contributions from the community! Whether it's bug reports, feature r
 - [General contribution guideline and flow](https://google.github.io/adk-docs/contributing-guide/).
 - Then if you want to contribute code, please read [Code Contributing Guidelines](./CONTRIBUTING.md) to get started.
 
+## Vibe Coding
+
+If you are to develop agent via vibe coding the [llms.txt](./llms.txt) and the [llms-full.txt](./llms-full.txt) can be used as context to LLM. While the former one is a summarized one and the later one has the full information in case your LLM has big enough context window.
+
 ## 📄 License
 
 This project is licensed under the Apache 2.0 License - see the [LICENSE](LICENSE) file for details.

contributing/adk_project_overview_and_architecture.md

@@ -28,20 +28,22 @@ Google Agent Development Kit (ADK) for Python
 
 Adhere to this structure for compatibility with ADK tooling.
 
-my_adk_project/ \
-└── src/ \
-    └── my_app/ \
-        ├── agents/ \
-        │   ├── my_agent/ \
+```
+my_adk_project/
+└── src/
+    └── my_app/
+        ├── agents/
+        │   ├── my_agent/
         │   │   ├── __init__.py   # Must contain: from. import agent \
         │   │   └── agent.py      # Must contain: root_agent = Agent(...) \
-        │   └── another_agent/ \
-        │       ├── __init__.py \
+        │   └── another_agent/
+        │       ├── __init__.py
         │       └── agent.py\
+```
 
 agent.py: Must define the agent and assign it to a variable named root_agent. This is how ADK's tools find it.
 
-__init__.py: In each agent directory, it must contain from. import agent to make the agent discoverable.
+`__init__.py`: In each agent directory, it must contain from. import agent to make the agent discoverable.
 
 ## Local Development & Debugging
 
@@ -108,4 +110,3 @@ Test Cases: Create JSON files with input and a reference (expected tool calls an
 Metrics: tool_trajectory_avg_score (does it use tools correctly?) and response_match_score (is the final answer good?).
 
 Run via: adk web (UI), pytest (for CI/CD), or adk eval (CLI).
-

contributing/samples/a2a_auth/agent.py

@@ -13,11 +13,11 @@
 # limitations under the License.
 
 
-from google.adk.agents import Agent
+from google.adk.agents.llm_agent import Agent
 from google.adk.agents.remote_a2a_agent import AGENT_CARD_WELL_KNOWN_PATH
 from google.adk.agents.remote_a2a_agent import RemoteA2aAgent
 from google.adk.tools.langchain_tool import LangchainTool
-from langchain_community.tools import YouTubeSearchTool
+from langchain_community.tools.youtube.search import YouTubeSearchTool
 
 # Instantiate the tool
 langchain_yt_tool = YouTubeSearchTool()

contributing/samples/a2a_basic/agent.py

@@ -14,7 +14,7 @@
 
 import random
 
-from google.adk.agents import Agent
+from google.adk.agents.llm_agent import Agent
 from google.adk.agents.remote_a2a_agent import AGENT_CARD_WELL_KNOWN_PATH
 from google.adk.agents.remote_a2a_agent import RemoteA2aAgent
 from google.adk.tools.example_tool import ExampleTool

contributing/samples/a2a_human_in_loop/agent.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 
 
-from google.adk import Agent
+from google.adk.agents.llm_agent import Agent
 from google.adk.agents.remote_a2a_agent import AGENT_CARD_WELL_KNOWN_PATH
 from google.adk.agents.remote_a2a_agent import RemoteA2aAgent
 from google.genai import types

contributing/samples/a2a_human_in_loop/remote_a2a/human_in_loop/agent.py

@@ -15,8 +15,8 @@
 from typing import Any
 
 from google.adk import Agent
-from google.adk.tools import ToolContext
 from google.adk.tools.long_running_tool import LongRunningFunctionTool
+from google.adk.tools.tool_context import ToolContext
 from google.genai import types
 
 

contributing/samples/a2a_root/README.md

@@ -0,0 +1,123 @@
+# A2A Root Sample Agent
+
+This sample demonstrates how to use a **remote Agent-to-Agent (A2A) agent as the root agent** in the Agent Development Kit (ADK). This is a simplified approach where the main agent is actually a remote A2A service, also showcasing how to run remote agents using uvicorn command.
+
+## Overview
+
+The A2A Root sample consists of:
+
+- **Root Agent** (`agent.py`): A remote A2A agent proxy as root agent that talks to a remote a2a agent running on a separate server
+- **Remote Hello World Agent** (`remote_a2a/hello_world/agent.py`): The actual agent implementation that handles dice rolling and prime number checking running on remote server
+
+## Architecture
+
+```
+┌─────────────────┐    ┌────────────────────┐
+│   Root Agent    │───▶│   Remote Hello     │
+│ (RemoteA2aAgent)│    │   World Agent      │
+│ (localhost:8000)│    │  (localhost:8001)  │
+└─────────────────┘    └────────────────────┘
+```
+
+## Key Features
+
+### 1. **Remote A2A as Root Agent**
+- The `root_agent` is a `RemoteA2aAgent` that connects to a remote A2A service
+- Demonstrates how to use remote agents as the primary agent instead of local agents
+- Shows the flexibility of the A2A architecture for distributed agent deployment
+
+### 2. **Uvicorn Server Deployment**
+- The remote agent is served using uvicorn, a lightweight ASGI server
+- Demonstrates a simple way to deploy A2A agents without using the ADK CLI
+- Shows how to expose A2A agents as standalone web services
+
+### 3. **Agent Functionality**
+- **Dice Rolling**: Can roll dice with configurable number of sides
+- **Prime Number Checking**: Can check if numbers are prime
+- **State Management**: Maintains roll history in tool context
+- **Parallel Tool Execution**: Can use multiple tools in parallel
+
+### 4. **Simple Deployment Pattern**
+- Uses the `to_a2a()` utility to convert a standard ADK agent to an A2A service
+- Minimal configuration required for remote agent deployment
+
+## Setup and Usage
+
+### Prerequisites
+
+1. **Start the Remote A2A Agent server**:
+   ```bash
+   # Start the remote agent using uvicorn
+   uvicorn contributing.samples.a2a_root.remote_a2a.hello_world.agent:a2a_app --host localhost --port 8001
+   ```
+
+2. **Run the Main Agent**:
+   ```bash
+   # In a separate terminal, run the adk web server
+   adk web contributing/samples/
+   ```
+
+### Example Interactions
+
+Once both services are running, you can interact with the root agent:
+
+**Simple Dice Rolling:**
+```
+User: Roll a 6-sided die
+Bot: I rolled a 4 for you.
+```
+
+**Prime Number Checking:**
+```
+User: Is 7 a prime number?
+Bot: Yes, 7 is a prime number.
+```
+
+**Combined Operations:**
+```
+User: Roll a 10-sided die and check if it's prime
+Bot: I rolled an 8 for you.
+Bot: 8 is not a prime number.
+```
+
+**Multiple Rolls with Prime Checking:**
+```
+User: Roll a die 3 times and check which results are prime
+Bot: I rolled a 3 for you.
+Bot: I rolled a 7 for you.
+Bot: I rolled a 4 for you.
+Bot: 3, 7 are prime numbers.
+```
+
+## Code Structure
+
+### Root Agent (`agent.py`)
+
+- **`root_agent`**: A `RemoteA2aAgent` that connects to the remote A2A service
+- **Agent Card URL**: Points to the well-known agent card endpoint on the remote server
+
+### Remote Hello World Agent (`remote_a2a/hello_world/agent.py`)
+
+- **`roll_die(sides: int)`**: Function tool for rolling dice with state management
+- **`check_prime(nums: list[int])`**: Async function for prime number checking
+- **`root_agent`**: The main agent with comprehensive instructions
+- **`a2a_app`**: The A2A application created using `to_a2a()` utility
+
+
+
+## Troubleshooting
+
+**Connection Issues:**
+- Ensure the uvicorn server is running on port 8001
+- Check that no firewall is blocking localhost connections
+- Verify the agent card URL in the root agent configuration
+- Check uvicorn logs for any startup errors
+
+**Agent Not Responding:**
+- Check the uvicorn server logs for errors
+- Verify the agent instructions are clear and unambiguous
+- Ensure the A2A app is properly configured with the correct port
+
+**Uvicorn Issues:**
+- Make sure the module path is correct: `contributing.samples.a2a_root.remote_a2a.hello_world.agent:a2a_app`
+- Check that all dependencies are installed

contributing/samples/a2a_root/agent.py

@@ -0,0 +1,24 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from a2a.utils.constants import AGENT_CARD_WELL_KNOWN_PATH
+from google.adk.agents.remote_a2a_agent import RemoteA2aAgent
+
+root_agent = RemoteA2aAgent(
+    name="hello_world_agent",
+    description=(
+        "Helpful assistant that can roll dice and check if numbers are prime."
+    ),
+    agent_card=f"http://localhost:8001/{AGENT_CARD_WELL_KNOWN_PATH}",
+)

contributing/samples/a2a_root/remote_a2a/hello_world/__init__.py

@@ -0,0 +1,15 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from . import agent

contributing/samples/a2a_root/remote_a2a/hello_world/agent.py

@@ -0,0 +1,111 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import random
+
+from google.adk import Agent
+from google.adk.a2a.utils.agent_to_a2a import to_a2a
+from google.adk.tools.tool_context import ToolContext
+from google.genai import types
+
+
+def roll_die(sides: int, tool_context: ToolContext) -> int:
+  """Roll a die and return the rolled result.
+
+  Args:
+    sides: The integer number of sides the die has.
+    tool_context: the tool context
+  Returns:
+    An integer of the result of rolling the die.
+  """
+  result = random.randint(1, sides)
+  if not 'rolls' in tool_context.state:
+    tool_context.state['rolls'] = []
+
+  tool_context.state['rolls'] = tool_context.state['rolls'] + [result]
+  return result
+
+
+async def check_prime(nums: list[int]) -> str:
+  """Check if a given list of numbers are prime.
+
+  Args:
+    nums: The list of numbers to check.
+
+  Returns:
+    A str indicating which number is prime.
+  """
+  primes = set()
+  for number in nums:
+    number = int(number)
+    if number <= 1:
+      continue
+    is_prime = True
+    for i in range(2, int(number**0.5) + 1):
+      if number % i == 0:
+        is_prime = False
+        break
+    if is_prime:
+      primes.add(number)
+  return (
+      'No prime numbers found.'
+      if not primes
+      else f"{', '.join(str(num) for num in primes)} are prime numbers."
+  )
+
+
+root_agent = Agent(
+    model='gemini-2.0-flash',
+    name='hello_world_agent',
+    description=(
+        'hello world agent that can roll a dice of 8 sides and check prime'
+        ' numbers.'
+    ),
+    instruction="""
+      You roll dice and answer questions about the outcome of the dice rolls.
+      You can roll dice of different sizes.
+      You can use multiple tools in parallel by calling functions in parallel(in one request and in one round).
+      It is ok to discuss previous dice roles, and comment on the dice rolls.
+      When you are asked to roll a die, you must call the roll_die tool with the number of sides. Be sure to pass in an integer. Do not pass in a string.
+      You should never roll a die on your own.
+      When checking prime numbers, call the check_prime tool with a list of integers. Be sure to pass in a list of integers. You should never pass in a string.
+      You should not check prime numbers before calling the tool.
+      When you are asked to roll a die and check prime numbers, you should always make the following two function calls:
+      1. You should first call the roll_die tool to get a roll. Wait for the function response before calling the check_prime tool.
+      2. After you get the function response from roll_die tool, you should call the check_prime tool with the roll_die result.
+        2.1 If user asks you to check primes based on previous rolls, make sure you include the previous rolls in the list.
+      3. When you respond, you must include the roll_die result from step 1.
+      You should always perform the previous 3 steps when asking for a roll and checking prime numbers.
+      You should not rely on the previous history on prime results.
+    """,
+    tools=[
+        roll_die,
+        check_prime,
+    ],
+    # planner=BuiltInPlanner(
+    #     thinking_config=types.ThinkingConfig(
+    #         include_thoughts=True,
+    #     ),
+    # ),
+    generate_content_config=types.GenerateContentConfig(
+        safety_settings=[
+            types.SafetySetting(  # avoid false alarm about rolling dice.
+                category=types.HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT,
+                threshold=types.HarmBlockThreshold.OFF,
+            ),
+        ]
+    ),
+)
+
+a2a_app = to_a2a(root_agent, port=8001)

contributing/samples/adk_answering_agent/README.md

@@ -2,7 +2,11 @@
 
 The ADK Answering Agent is a Python-based agent designed to help answer questions in GitHub discussions for the `google/adk-python` repository. It uses a large language model to analyze open discussions, retrieve information from document store, generate response, and post a comment in the github discussion.
 
-This agent can be operated in three distinct modes: an interactive mode for local use, a batch script mode for oncall use, or as a fully automated GitHub Actions workflow (TBD).
+This agent can be operated in three distinct modes:
+
+- An interactive mode for local use.
+- A batch script mode for oncall use.
+- A fully automated GitHub Actions workflow (TBD).
 
 ---
 
@@ -50,6 +54,15 @@ The `main.py` is reserved for the Github Workflow. The detailed setup for the au
 
 ---
 
+## Update the Knowledge Base
+
+The `upload_docs_to_vertex_ai_search.py` is a script to upload ADK related docs to Vertex AI Search datastore to update the knowledge base. It can be executed with the following command in your terminal:
+
+```bash
+export PYTHONPATH=contributing/samples # If not already exported
+python -m adk_answering_agent.upload_docs_to_vertex_ai_search
+```
+
 ## Setup and Configuration
 
 Whether running in interactive or workflow mode, the agent requires the following setup.
@@ -59,7 +72,7 @@ The agent requires the following Python libraries.
 
 ```bash
 pip install --upgrade pip
-pip install google-adk requests
+pip install google-adk
 ```
 
 The agent also requires gcloud login:
@@ -68,6 +81,12 @@ The agent also requires gcloud login:
 gcloud auth application-default login
 ```
 
+The upload script requires the following additional Python libraries.
+
+```bash
+pip install google-cloud-storage google-cloud-discoveryengine
+```
+
 ### Environment Variables
 The following environment variables are required for the agent to connect to the necessary services.
 
@@ -75,9 +94,15 @@ The following environment variables are required for the agent to connect to the
 * `GOOGLE_GENAI_USE_VERTEXAI=TRUE`: **(Required)** Use Google Vertex AI for the authentication.
 * `GOOGLE_CLOUD_PROJECT=YOUR_PROJECT_ID`: **(Required)** The Google Cloud project ID.
 * `GOOGLE_CLOUD_LOCATION=LOCATION`: **(Required)** The Google Cloud region.
-* `VERTEXAI_DATASTORE_ID=YOUR_DATASTORE_ID`: **(Required)** The Vertex AI datastore ID for the document store (i.e. knowledge base).
+* `VERTEXAI_DATASTORE_ID=YOUR_DATASTORE_ID`: **(Required)** The full Vertex AI datastore ID for the document store (i.e. knowledge base), with the format of `projects/{project_number}/locations/{location}/collections/{collection}/dataStores/{datastore_id}`.
 * `OWNER`: The GitHub organization or username that owns the repository (e.g., `google`). Needed for both modes.
 * `REPO`: The name of the GitHub repository (e.g., `adk-python`). Needed for both modes.
 * `INTERACTIVE`: Controls the agent's interaction mode. For the automated workflow, this is set to `0`. For interactive mode, it should be set to `1` or left unset.
 
+The following environment variables are required to upload the docs to update the knowledge base.
+
+* `GCS_BUCKET_NAME=YOUR_GCS_BUCKET_NAME`: **(Required)** The name of the GCS bucket to store the documents.
+* `ADK_DOCS_ROOT_PATH=YOUR_ADK_DOCS_ROOT_PATH`: **(Required)** Path to the root of the downloaded adk-docs repo.
+* `ADK_PYTHON_ROOT_PATH=YOUR_ADK_PYTHON_ROOT_PATH`: **(Required)** Path to the root of the downloaded adk-python repo.
+
 For local execution in interactive mode, you can place these variables in a `.env` file in the project's root directory. For the GitHub workflow, they should be configured as repository secrets.

contributing/samples/adk_answering_agent/agent.py

@@ -21,8 +21,8 @@ from adk_answering_agent.settings import REPO
 from adk_answering_agent.settings import VERTEXAI_DATASTORE_ID
 from adk_answering_agent.utils import error_response
 from adk_answering_agent.utils import run_graphql_query
-from google.adk.agents import Agent
-from google.adk.tools import VertexAiSearchTool
+from google.adk.agents.llm_agent import Agent
+from google.adk.tools.vertex_ai_search_tool import VertexAiSearchTool
 import requests
 
 if IS_INTERACTIVE:

contributing/samples/adk_answering_agent/settings.py

@@ -29,9 +29,14 @@ VERTEXAI_DATASTORE_ID = os.getenv("VERTEXAI_DATASTORE_ID")
 if not VERTEXAI_DATASTORE_ID:
   raise ValueError("VERTEXAI_DATASTORE_ID environment variable not set")
 
+GOOGLE_CLOUD_PROJECT = os.getenv("GOOGLE_CLOUD_PROJECT")
+GCS_BUCKET_NAME = os.getenv("GCS_BUCKET_NAME")
+ADK_DOCS_ROOT_PATH = os.getenv("ADK_DOCS_ROOT_PATH")
+ADK_PYTHON_ROOT_PATH = os.getenv("ADK_PYTHON_ROOT_PATH")
+
 OWNER = os.getenv("OWNER", "google")
 REPO = os.getenv("REPO", "adk-python")
-BOT_RESPONSE_LABEL = os.getenv("BOT_RESPONSE_LABEL", "bot_responded")
+BOT_RESPONSE_LABEL = os.getenv("BOT_RESPONSE_LABEL", "bot responded")
 DISCUSSION_NUMBER = os.getenv("DISCUSSION_NUMBER")
 
 IS_INTERACTIVE = os.getenv("INTERACTIVE", "1").lower() in ["true", "1"]

contributing/samples/adk_answering_agent/upload_docs_to_vertex_ai_search.py

@@ -0,0 +1,222 @@
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import sys
+
+from adk_answering_agent.settings import ADK_DOCS_ROOT_PATH
+from adk_answering_agent.settings import ADK_PYTHON_ROOT_PATH
+from adk_answering_agent.settings import GCS_BUCKET_NAME
+from adk_answering_agent.settings import GOOGLE_CLOUD_PROJECT
+from adk_answering_agent.settings import VERTEXAI_DATASTORE_ID
+from google.api_core.exceptions import GoogleAPICallError
+from google.cloud import discoveryengine_v1beta as discoveryengine
+from google.cloud import storage
+import markdown
+
+GCS_PREFIX_TO_ROOT_PATH = {
+    "adk-docs": ADK_DOCS_ROOT_PATH,
+    "adk-python": ADK_PYTHON_ROOT_PATH,
+}
+
+
+def cleanup_gcs_prefix(project_id: str, bucket_name: str, prefix: str) -> bool:
+  """Delete all the objects with the given prefix in the bucket."""
+  print(f"Start cleaning up GCS: gs://{bucket_name}/{prefix}...")
+  try:
+    storage_client = storage.Client(project=project_id)
+    bucket = storage_client.bucket(bucket_name)
+    blobs = list(bucket.list_blobs(prefix=prefix))
+
+    if not blobs:
+      print("GCS target location is already empty, no need to clean up.")
+      return True
+
+    bucket.delete_blobs(blobs)
+    print(f"Successfully deleted {len(blobs)} objects.")
+    return True
+  except GoogleAPICallError as e:
+    print(f"[ERROR] Failed to clean up GCS: {e}", file=sys.stderr)
+    return False
+
+
+def upload_directory_to_gcs(
+    source_directory: str, project_id: str, bucket_name: str, prefix: str
+) -> bool:
+  """Upload the whole directory into GCS."""
+  print(
+      f"Start uploading directory {source_directory} to GCS:"
+      f" gs://{bucket_name}/{prefix}..."
+  )
+
+  if not os.path.isdir(source_directory):
+    print(f"[Error] {source_directory} is not a directory or does not exist.")
+    return False
+
+  storage_client = storage.Client(project=project_id)
+  bucket = storage_client.bucket(bucket_name)
+  file_count = 0
+  for root, dirs, files in os.walk(source_directory):
+    # Modify the 'dirs' list in-place to prevent os.walk from descending
+    # into hidden directories.
+    dirs[:] = [d for d in dirs if not d.startswith(".")]
+
+    # Keep only .md and .py files.
+    files = [f for f in files if f.endswith(".md") or f.endswith(".py")]
+
+    for filename in files:
+      local_path = os.path.join(root, filename)
+
+      relative_path = os.path.relpath(local_path, source_directory)
+      gcs_path = os.path.join(prefix, relative_path)
+
+      try:
+        content_type = None
+        if filename.lower().endswith(".md"):
+          # Vertex AI search doesn't recognize text/markdown,
+          # convert it to html and use text/html instead
+          content_type = "text/html"
+          with open(local_path, "r", encoding="utf-8") as f:
+            md_content = f.read()
+          html_content = markdown.markdown(
+              md_content, output_format="html5", encoding="utf-8"
+          )
+          if not html_content:
+            print("  - Skipped empty file: " + local_path)
+            continue
+          gcs_path = gcs_path.removesuffix(".md") + ".html"
+          bucket.blob(gcs_path).upload_from_string(
+              html_content, content_type=content_type
+          )
+        else:  # Python files
+          bucket.blob(gcs_path).upload_from_filename(
+              local_path, content_type=content_type
+          )
+        type_msg = (
+            f"(type {content_type})" if content_type else "(type auto-detect)"
+        )
+        print(
+            f"  - Uploaded {type_msg}: {local_path} ->"
+            f" gs://{bucket_name}/{gcs_path}"
+        )
+        file_count += 1
+      except GoogleAPICallError as e:
+        print(
+            f"[ERROR] Error uploading file {local_path}: {e}", file=sys.stderr
+        )
+        return False
+
+  print(f"Sucessfully uploaded {file_count} files to GCS.")
+  return True
+
+
+def import_from_gcs_to_vertex_ai(
+    full_datastore_id: str,
+    gcs_bucket: str,
+) -> bool:
+  """Triggers a bulk import task from a GCS folder to Vertex AI Search."""
+  print(f"Triggering FULL SYNC import from gs://{gcs_bucket}/**...")
+
+  try:
+    client = discoveryengine.DocumentServiceClient()
+    gcs_uri = f"gs://{gcs_bucket}/**"
+    request = discoveryengine.ImportDocumentsRequest(
+        # parent has the format of
+        # "projects/{project_number}/locations/{location}/collections/{collection}/dataStores/{datastore_id}/branches/default_branch"
+        parent=full_datastore_id + "/branches/default_branch",
+        # Specify the GCS source and use "content" for unstructed data.
+        gcs_source=discoveryengine.GcsSource(
+            input_uris=[gcs_uri], data_schema="content"
+        ),
+        reconciliation_mode=discoveryengine.ImportDocumentsRequest.ReconciliationMode.FULL,
+    )
+    operation = client.import_documents(request=request)
+    print(
+        "Successfully started full sync import operation."
+        f"Operation Name: {operation.operation.name}"
+    )
+    return True
+
+  except GoogleAPICallError as e:
+    print(f"[ERROR] Error triggering import: {e}", file=sys.stderr)
+    return False
+
+
+def main():
+  # Check required environment variables.
+  if not GOOGLE_CLOUD_PROJECT:
+    print(
+        "[ERROR] GOOGLE_CLOUD_PROJECT environment variable not set. Exiting...",
+        file=sys.stderr,
+    )
+    return 1
+  if not GCS_BUCKET_NAME:
+    print(
+        "[ERROR] GCS_BUCKET_NAME environment variable not set. Exiting...",
+        file=sys.stderr,
+    )
+    return 1
+  if not VERTEXAI_DATASTORE_ID:
+    print(
+        "[ERROR] VERTEXAI_DATASTORE_ID environment variable not set."
+        " Exiting...",
+        file=sys.stderr,
+    )
+    return 1
+  if not ADK_DOCS_ROOT_PATH:
+    print(
+        "[ERROR] ADK_DOCS_ROOT_PATH environment variable not set. Exiting...",
+        file=sys.stderr,
+    )
+    return 1
+  if not ADK_PYTHON_ROOT_PATH:
+    print(
+        "[ERROR] ADK_PYTHON_ROOT_PATH environment variable not set. Exiting...",
+        file=sys.stderr,
+    )
+    return 1
+
+  for gcs_prefix in GCS_PREFIX_TO_ROOT_PATH:
+    # 1. Cleanup the GSC for a clean start.
+    if not cleanup_gcs_prefix(
+        GOOGLE_CLOUD_PROJECT, GCS_BUCKET_NAME, gcs_prefix
+    ):
+      print("[ERROR] Failed to clean up GCS. Exiting...", file=sys.stderr)
+      return 1
+
+    # 2. Upload the docs to GCS.
+    if not upload_directory_to_gcs(
+        GCS_PREFIX_TO_ROOT_PATH[gcs_prefix],
+        GOOGLE_CLOUD_PROJECT,
+        GCS_BUCKET_NAME,
+        gcs_prefix,
+    ):
+      print("[ERROR] Failed to upload docs to GCS. Exiting...", file=sys.stderr)
+      return 1
+
+  # 3. Import the docs from GCS to Vertex AI Search.
+  if not import_from_gcs_to_vertex_ai(VERTEXAI_DATASTORE_ID, GCS_BUCKET_NAME):
+    print(
+        "[ERROR] Failed to import docs from GCS to Vertex AI Search."
+        " Exiting...",
+        file=sys.stderr,
+    )
+    return 1
+
+  print("--- Sync task has been successfully initiated ---")
+  return 0
+
+
+if __name__ == "__main__":
+  sys.exit(main())

contributing/samples/adk_pr_agent/main.py

@@ -20,7 +20,7 @@ import time
 import agent
 from google.adk.agents.run_config import RunConfig
 from google.adk.runners import InMemoryRunner
-from google.adk.sessions import Session
+from google.adk.sessions.session import Session
 from google.genai import types
 
 

contributing/samples/adk_pr_triaging_agent/README.md

@@ -0,0 +1,68 @@
+# ADK Pull Request Triaging Assistant
+
+The ADK Pull Request (PR) Triaging Assistant is a Python-based agent designed to help manage and triage GitHub pull requests for the `google/adk-python` repository. It uses a large language model to analyze new and unlabelled pull requests, recommend appropriate labels, assign a reviewer, and check contribution guides based on a predefined set of rules.
+
+This agent can be operated in two distinct modes:
+
+* an interactive mode for local use
+* a fully automated GitHub Actions workflow.
+
+---
+
+## Interactive Mode
+
+This mode allows you to run the agent locally to review its recommendations in real-time before any changes are made to your repository's pull requests.
+
+### Features
+* **Web Interface**: The agent's interactive mode can be rendered in a web browser using the ADK's `adk web` command.
+* **User Approval**: In interactive mode, the agent is instructed to ask for your confirmation before applying a label or posting a comment to a GitHub pull request.
+
+### Running in Interactive Mode
+To run the agent in interactive mode, first set the required environment variables. Then, execute the following command in your terminal:
+
+```bash
+adk web
+```
+This will start a local server and provide a URL to access the agent's web interface in your browser.
+
+---
+
+## GitHub Workflow Mode
+
+For automated, hands-off PR triaging, the agent can be integrated directly into your repository's CI/CD pipeline using a GitHub Actions workflow.
+
+### Workflow Triggers
+The GitHub workflow is configured to run on specific triggers:
+
+*  **Pull Request Events**: The workflow executes automatically whenever a new PR is `opened` or an existing one is `reopened` or `edited`.
+
+### Automated Labeling
+When running as part of the GitHub workflow, the agent operates non-interactively. It identifies and applies the best label or posts a comment directly without requiring user approval. This behavior is configured by setting the `INTERACTIVE` environment variable to `0` in the workflow file.
+
+### Workflow Configuration
+The workflow is defined in a YAML file (`.github/workflows/pr-triage.yml`). This file contains the steps to check out the code, set up the Python environment, install dependencies, and run the triaging script with the necessary environment variables and secrets.
+
+---
+
+## Setup and Configuration
+
+Whether running in interactive or workflow mode, the agent requires the following setup.
+
+### Dependencies
+The agent requires the following Python libraries.
+
+```bash
+pip install --upgrade pip
+pip install google-adk
+```
+
+### Environment Variables
+The following environment variables are required for the agent to connect to the necessary services.
+
+* `GITHUB_TOKEN`: **(Required)** A GitHub Personal Access Token with `pull_requests:write` permissions. Needed for both interactive and workflow modes.
+* `GOOGLE_API_KEY`: **(Required)** Your API key for the Gemini API. Needed for both interactive and workflow modes.
+* `OWNER`: The GitHub organization or username that owns the repository (e.g., `google`). Needed for both modes.
+* `REPO`: The name of the GitHub repository (e.g., `adk-python`). Needed for both modes.
+* `INTERACTIVE`: Controls the agent's interaction mode. For the automated workflow, this is set to `0`. For interactive mode, it should be set to `1` or left unset.
+
+For local execution in interactive mode, you can place these variables in a `.env` file in the project's root directory. For the GitHub workflow, they should be configured as repository secrets.