Clean up runner resources in AgentTool after sub-agent execution to ensure
MCP sessions are closed in the correct async task context. Without this fix,
MCP sessions were cleaned up during garbage collection in a different task,
causing "Attempted to exit cancel scope in a different task" errors.
This fix ensures that `runner.close()` is called immediately after the
sub-agent finishes executing, properly closing all MCP sessions and other
resources within the same async task context they were created in.
Also adds two demo agents showing how to use AgentTool with MCP tools:
- mcp_in_agent_tool_remote: Uses SSE mode (remote server connection)
- mcp_in_agent_tool_stdio: Uses stdio mode (local subprocess)
Both demos use Gemini 2.5 Flash and include zero-installation setup using uvx.
Related: #1112, #929
The blob content is often large and binary, which makes the logs unreadable and can cause excessive logging.
Co-authored-by: Hangfei Lin <hangfei@google.com>
PiperOrigin-RevId: 828523413
The sample agent now uses updated model names for Gemini Live, including a new Vertex model as the default and a new AI Studio model option.
Co-authored-by: Hangfei Lin <hangfei@google.com>
PiperOrigin-RevId: 828515811
Merge https://github.com/google/adk-python/pull/3381
### Link to Issue or Description of Change
**1. Link to an existing issue (if applicable):**
- Closes: #3363
- This PR sets a max column width for the table printed in detailed output of agent evaluations.
**Problem:**
The detailed output of agent evaluations is not readable due to rows in the table getting wrapped. This happens when there are long text values in cells.
<img width="1904" height="717" alt="508807185-9e8fe1c3-d04a-43dd-acf9-0befaa1b247d" src="https://github.com/user-attachments/assets/61526ad2-8a9e-4c18-83e2-51a3b9b32d2b" />
**Solution:**
Existing code uses `tabulate` python package to format the table. We can set a maximum column width using `maxcolwidths` parameter. I have set it to `25`.
After the fix:
<img width="1882" height="711" alt="508810179-b91c5bca-fb43-480b-90ff-bca2e909417c" src="https://github.com/user-attachments/assets/b653f825-719e-4101-9acb-e28a52694cf8" />
### Testing Plan
I have manually tested if the output is properly displayed after changes. Please let me know if any unit tests can be added for this.
**Unit Tests:**
- [ ] I have added or updated unit tests for my change.
- [x] All unit tests pass locally.
<img width="1627" height="39" alt="image" src="https://github.com/user-attachments/assets/59a70619-3669-4113-8ab7-dcff130ee241" />
**Manual End-to-End (E2E) Tests:**
1. Create a simple agent using adk (preferably an agent that outputs a long text).
2. Create an evalset for this agent.
3. Run the evalset with `print_detailed_results` option and check if the output is properly displayed.
If you want a quick setup for testing this, I have a sample repo with an agent and an evalset [here](https://github.com/nimanthadilz/adk-test/tree/reproduce-print-detailed-results). You will have to manually build & install the fixed adk version to test it.
### Checklist
- [x] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document.
- [x] I have performed a self-review of my own code.
- [x] 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.
- [x] New and existing unit tests pass locally with my changes.
- [x] I have manually tested my changes end-to-end.
- [x] Any dependent changes have been merged and published in downstream modules.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3381 from nimanthadilz:fix-eval-output-rows-wrapping-issue f6d40121f621df60c4596a1c62e0c54e4da309d3
PiperOrigin-RevId: 828265715
Merge https://github.com/google/adk-python/pull/2651
### Summary
Correct a misspelling in the build configuration:
- "swtich" → "switch" in `pyproject.toml`.
### Rationale
This is a spelling fix only. It improves readability and avoids potential confusion in configuration.
There is no impact on runtime behavior, tests, or public APIs.
### Notes
- Follows Conventional Commits style for build/config changes (`build:`).
- CLA status should be green via the Google CLA bot.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2651 from marsboy02:docs/fix-type-pyproject b78c014c864b1a143ffc157c7a8c807f5f19d31d
PiperOrigin-RevId: 828221776
This change introduces BigQueryLoggerConfig to allow customization of the BigQueryAgentAnalyticsPlugin. Users can now enable/disable the plugin, specify event type allowlists and denylists, and provide a custom function to format or redact the content field before logging to BigQuery. The content logged for model and tool errors has also been enhanced.
PiperOrigin-RevId: 828172241
Add support for MCP prompts via the McpInstructionProvider class, which can be specified as an agent's instruction.
Co-authored-by: Kathy Wu <wukathy@google.com>
PiperOrigin-RevId: 828166051
Users were getting spammed with this log even though their tools didn't require authentication. To fix, reduce the log level to DEBUG so that it doesn't show up by default.
Co-authored-by: Kathy Wu <wukathy@google.com>
PiperOrigin-RevId: 828161281
This change introduces a new section in the README.md to highlight the `adk-python-community` GitHub repository, describing it as a place for community-contributed tools and integrations.
Co-authored-by: Hangfei Lin <hangfei@google.com>
PiperOrigin-RevId: 828155205
Populate the usage_metadata field for live events with the metadata provided by the Gemini live API.
Co-authored-by: Kathy Wu <wukathy@google.com>
PiperOrigin-RevId: 828124232
This lets users to specify `drop_params` when initializing `LiteLlm`, which will be forwarded to LiteLLM's `acompletion` or `completion` calls
Close#1718
Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 828058105
LiteLLM providers can extract the MIME type from the data URI. Removing the separate `format` field avoids redundancy and potential issues with backends that may reject requests containing this field.
Close#2017
Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 828014286
ARIMA supports both historical data and future data anomaly detection. This CL add how the tool support future table anomaly detection.
PiperOrigin-RevId: 827803748
Merge https://github.com/google/adk-python/pull/3365
**Please ensure you have read the [contribution guide](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) before creating a pull request.**
fix typo for several files.
### Checklist
- [x] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document.
- [x] 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.
- [x] New and existing unit tests pass locally with my changes.
- [x] I have manually tested my changes end-to-end.
- [x] Any dependent changes have been merged and published in downstream modules.
### Additional context
_Add any other context or screenshots about the feature request here._
Co-authored-by: Liang Wu <wuliang@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3365 from UlookEE:fix_typo 1469de4ea354d1c205268b999183ee86c9d6a1d5
PiperOrigin-RevId: 827724001
Merge #3163
END_PUBLIC
Hello,
Since global_instruction has been deprecated, I’m migrating to GlobalInstructionPlugin.
During the migration, I encountered an error and am submitting this PR to fix it.
In [df05ed6](https://github.com/google/adk-python/commit/df05ed6b3b7b218d85fddc1acd6617802cdf6f2a) ,
GlobalInstructionPlugin references invocation_context, but CallbackContext actually contains _invocation_context.
This mismatch always causes an error during execution.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3180 from UlookEE:fix_global_instruction_plugin e289a12d69812f0abcfe77db0114fdb2045b31bc
PiperOrigin-RevId: 827682501
Merge https://github.com/google/adk-python/pull/2326
`adk run --help` (adk 1.9.0)
```
--resume FILE The json file that contains a previously saved session
(by--save_session option). The previous session will be
re-displayed. And user can continue to interact with the
agent.
```
## testing plan
N/A (because this is a simple string correction)
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2326 from ftnext:fix-typo-run-help-resume a896fa38e223b13e7edd8125d7b38139f1ca3712
PiperOrigin-RevId: 827311506
This fixes MCP authentication for gcloud service accounts. Previously it was failing to authenticate tool calls.
Co-authored-by: Kathy Wu <wukathy@google.com>
PiperOrigin-RevId: 826639044
Merge https://github.com/google/adk-python/pull/3345
Add run_debug() helper method to InMemoryRunner that reduces agent execution boilerplate from 7-8 lines to just 2 lines, making it ideal for quick experimentation, notebooks, and getting started with ADK.
**Key changes:**
• Introduce run_debug() to reduce boilerplate from 7-8 lines to 2 lines
• Enable quick testing in notebooks, REPL, and during development
• Support single or multiple messages with automatic session management
• Add verbose flag to show/hide tool calls and intermediate processing
• Add quiet flag to suppress console output while capturing events
• Extract event printing logic to reusable utility (utils/_debug_output.py)
• Include comprehensive test suite with 21 test cases covering all part types
• Provide complete working example with 8 usage patterns
• **This is a convenience method for experimentation, not a replacement for run_async()**
### Link to Issue or Description of Change
**1. Link to an existing issue (if applicable):**
* N/A - New feature to improve developer experience
**2. Or, if no issue exists, describe the change:**
**Problem:**
Developers need to write 7-8 lines of boilerplate code just to test a simple agent interaction during development. This creates friction for:
* New developers getting started with ADK
* Quick experimentation in Jupyter notebooks or Python REPL
* Debugging agent behavior during development
* Writing examples and tutorials
* Rapid prototyping of agent capabilities
**Solution:**
Introduce `run_debug()` as a convenience helper method specifically designed for quick experimentation and getting started scenarios. This method:
* **Is NOT a replacement for `run_async()`** - it's a developer convenience tool
* **Reduces boilerplate** from 7-8 lines to just 2 lines for simple testing
* **Handles session management automatically** with sensible defaults
* **Provides debugging visibility** with optional verbose flag for tool calls
* **Supports common patterns** like multiple messages and event capture
* **Type-safe implementation** using direct attribute access instead of getattr()
### Before vs After Comparison
**BEFORE - Current approach requires 7-8 lines of boilerplate:**
```python
from google.adk import Agent
from google.adk.runners import Runner
from google.adk.sessions import InMemorySessionService
from google.genai import types
# Define a simple agent
agent = Agent(
model="gemini-2.5-flash",
instruction="You are a helpful assistant"
)
# Need all this boilerplate just to test the agent
APP_NAME = "default"
USER_ID = "default"
session_service = InMemorySessionService()
runner = Runner(agent=agent, app_name=APP_NAME, session_service=session_service)
session = await session_service.create_session(
app_name=APP_NAME, user_id=USER_ID, session_id="default"
)
content = types.Content(role="user", parts=[types.Part.from_text("Hello")])
async for event in runner.run_async(
user_id=USER_ID, session_id=session.id, new_message=content
):
if event.content and event.content.parts:
print(event.content.parts[0].text)
```
**AFTER - With run_debug() helper, just 2 lines:**
```python
from google.adk import Agent
from google.adk.runners import InMemoryRunner
# Define the same agent
agent = Agent(
model="gemini-2.5-flash",
instruction="You are a helpful assistant"
)
# Test it with just 2 lines!
runner = InMemoryRunner(agent=agent)
await runner.run_debug("Hello")
```
### API Design
```python
async def run_debug(
self,
user_messages: str | list[str],
*,
user_id: str = 'debug_user_id',
session_id: str = 'debug_session_id',
run_config: RunConfig | None = None,
quiet: bool = False,
verbose: bool = False,
) -> list[Event]:
```
**Parameters:**
* `user_messages`: Single message string or list of messages (required)
* `user_id`: User identifier (default: 'debug_user_id')
* `session_id`: Session identifier for conversation continuity (default: 'debug_session_id')
* `run_config`: Optional advanced configuration
* `quiet`: Suppress console output (default: False)
* `verbose`: Show detailed tool calls and responses (default: False)
**Key Features:**
* **Always returns events** - Simplifies API, no conditional return type
* **Type-safe implementation** - Uses direct attribute access on Pydantic models
* **Text buffering** - Consecutive text parts printed without repeated author prefix
* **Smart truncation** - Long tool args/responses truncated for readability
* **Clean session management** - Get-then-create pattern, no try/except
* **Reusable printing logic** - Extracted to utils/_debug_output.py for other tools
### Implementation Highlights
**1. Event Printing Utility (utils/_debug_output.py):**
* Modular print_event() function for displaying events
* Text buffering to combine consecutive text parts
* Configurable truncation for different content types:
- Function args: 50 chars max
- Function responses: 100 chars max
- Code output: 100 chars max
* Supports all ADK part types (text, function_call, executable_code, inline_data, file_data)
**2. Session Management:**
```python
# Clean get-then-create pattern (no try/except)
session = await self.session_service.get_session(
app_name=self.app_name, user_id=user_id, session_id=session_id
)
if not session:
session = await self.session_service.create_session(
app_name=self.app_name, user_id=user_id, session_id=session_id
)
```
**3. Type-Safe Event Processing:**
* Direct attribute access on Pydantic models (no getattr() or hasattr())
* Proper handling of all part types
* Leverages `from __future__ import annotations` for duck typing
### Important Note on Scope
`run_debug()` is a **convenience method for experimentation only**. For production applications requiring:
* Custom session services (Spanner, Cloud SQL)
* Fine-grained event processing control
* Error recovery and resumability
* Performance optimization
* Complex authentication flows
Continue using the standard `run_async()` method. The `run_debug()` helper is specifically designed to lower the barrier to entry and speed up the development/testing cycle.
### Testing Plan
**Unit Tests (21 test cases in tests/unittests/runners/test_runner_debug.py):**
**Core functionality (7 tests):**
* âś… Single message execution and event return
* âś… Multiple messages in sequence
* âś… Quiet mode (suppresses output)
* âś… Custom session_id configuration
* âś… Custom user_id configuration
* âś… RunConfig passthrough
* âś… Session persistence across calls
**Part type handling (8 tests):**
* âś… Tool calls and responses (verbose mode)
* âś… Executable code parts
* âś… Code execution result parts
* âś… Inline data (images)
* âś… File data references
* âś… Mixed part types in single event
* âś… Long output truncation
* âś… Verbose flag behavior (show/hide tools)
**Edge cases (6 tests):**
* âś… None text filtering
* âś… Existing session handling
* âś… Empty parts list
* âś… None event content
* âś… Verbose=False hides tool calls
* âś… Verbose=True shows tool calls
**All 21 tests passing in 3.8s** âś“
**Manual End-to-End (E2E) Tests:**
Tested all 8 example patterns in contributing/samples/runner_debug_example/main.py:
1. âś… Minimal 2-line usage
2. âś… Multiple sequential messages
3. âś… Session persistence across calls
4. âś… Multiple user sessions (Alice & Bob)
5. âś… Verbose mode for tool visibility
6. âś… Event capture with quiet mode
7. âś… Custom RunConfig integration
8. âś… Before/after comparison
### Files Changed
**Core implementation:**
* src/google/adk/runners.py - Added run_debug() method (~60 lines)
* src/google/adk/utils/_debug_output.py - Event printing utility (~106 lines)
**Tests:**
* tests/unittests/runners/test_runner_debug.py - Comprehensive test suite (21 tests)
**Examples:**
* contributing/samples/runner_debug_example/agent.py - Sample agent with tools
* contributing/samples/runner_debug_example/main.py - 8 usage examples
* contributing/samples/runner_debug_example/README.md - Complete documentation
### Checklist
- [x] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document
- [x] I have performed a self-review of my own code
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have added tests that prove my fix is effective or that my feature works
- [x] New and existing unit tests pass locally with my changes (21/21 passing)
- [x] I have manually tested my changes end-to-end (8 examples tested)
- [x] Code follows ADK style guide (relative imports, type hints, 2-space indentation)
- [x] Ran ./autoformat.sh before committing
- [x] Any dependent changes have been merged and published in downstream modules
### Additional Context
**Example with Tools (verbose mode):**
```python
# Create agent with tools
agent = Agent(
model="gemini-2.5-flash",
instruction="You can check weather and do calculations",
tools=[get_weather, calculate]
)
# Test with verbose to see tool calls
runner = InMemoryRunner(agent=agent)
await runner.run_debug("What's the weather in SF?", verbose=True)
# Output:
# User > What's the weather in SF?
# agent > [Calling tool: get_weather({'city': 'San Francisco'})]
# agent > [Tool result: {'result': 'Foggy, 15°C (59°F)'}]
# agent > The weather in San Francisco is foggy, 15°C (59°F).
```
**Complete Example Included:**
The PR includes a full working example in `contributing/samples/runner_debug_example/` with:
* Agent with weather and calculator tools
* 8 different usage patterns
* Comprehensive README with troubleshooting
* Safe AST-based expression evaluation
**Breaking Changes:** None - this is purely additive.
**Security:** Example uses AST-based expression evaluation instead of eval().
**Code Quality:**
* Type-safe implementation (no getattr() or hasattr())
* Modular design (printing logic separated into utility)
* Follows ADK conventions (relative imports, from __future__ import annotations)
* Comprehensive error handling (gracefully handles None content, empty parts)
* Well-documented with docstrings and inline comments
END_PUBLIC
```
---
## Key Changes from Original:
1. ✅ Updated parameter name: `user_queries` → `user_messages`
2. ✅ Updated parameter name: `session_name` → `session_id`
3. ✅ Updated parameter name: `print_output` → `quiet`
4. âś… Removed `return_events` parameter
5. ✅ Updated test count: 23 → 21
6. ✅ Changed "queries" → "messages" throughout
7. âś… Added implementation highlights section
8. âś… Added details about utils/_debug_output.py
9. âś… Updated default values to debug_user_id/debug_session_id
10. âś… Noted type-safe implementation
11. âś… Added Code Quality section
12. âś… Updated API signature to match final refactored version
13. âś… Removed optional return type (always returns list[Event])
Co-authored-by: Wei Sun (Jack) <weisun@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3345 from lavinigam-gcp:adk-runner-helper e0050b9f152d0f0e49e6501610d2c59a754fc571
PiperOrigin-RevId: 826607817
Removes the dataset_id field from the BigQuery table schema and from each log entry created by the BigQueryAgentAnalyticsPlugin. This field is redundant, as all rows logged to a specific table belong to the same dataset.
To ensure the plugin can still target the correct dataset, dataset_id is now a required argument in the BigQueryAgentAnalyticsPlugin constructor, and its default value has been removed.
The BigQuery client user_agent is also updated with plugin version info to help identify traffic originating from this plugin. Unit tests have been updated to reflect the removal of dataset_id from log entries.
PiperOrigin-RevId: 826596499
This change adds an optional `runner` parameter to the `to_a2a` function, enabling users to provide a pre-configured `Runner` instance instead of always using the default in-memory services. A new test case has been added to verify this functionality.
closes#3104
Co-authored-by: Dongyu Jia <dongyuj@google.com>
PiperOrigin-RevId: 826526861
This change introduces a sample (`migration_session_db`) demonstrating how to load a session database created with an older version of ADK (e.g., 1.15.0) and make it compatible with the current version. It includes a script (`db_migration.sh`) to alter the SQLite schema automatically. to_event is updated to handle potential discrepancies in pickled `EventActions` by using `model_copy` to ensure compatibility with the latest `EventActions` model definition.
Related to #3272#3197, Closes#3197#3272
Co-authored-by: Dongyu Jia <dongyuj@google.com>
PiperOrigin-RevId: 826524368
It is common for expected response and expected tool calls column to be empty for user simulated conversations. So, we don't render those.
Co-authored-by: Ankur Sharma <ankusharma@google.com>
PiperOrigin-RevId: 826202867
Previously we only do a simple prefix string matching, thus `agent_00` will match with `agent_0`
With this new change, we either check directly equality, or must expect seeing `agent_0.`. See added test for branches we now match / skip.
TBF `.` is also not a perfect delimiter (I would imagine users might put dot in agent names). We might consider a follow up that bans such agent names.
Tested with script in the linked issue (I updated prompt so we see which agent they see from):
Before:
```
[agent_8]: 73
[agent_0]: 97
[agent_1]: 73
[agent_5]: 97
[agent_4]: 73
[agent_2]: 73
[agent_3]: 73
[agent_9]: 93
[agent_6]: 73
[agent_7]: 1
[agent_70]: 1 (agent_7)
[agent_20]: 73 (agent_2)
[agent_30]: 73 (agent_3)
[agent_00]: 97 (agent_0)
[agent_40]: 73 (agent_4)
[agent_80]: 73 (agent_8)
[agent_50]: 97 (agent_5)
[agent_90]: 93 (agent_9)
[agent_10]: 73 (agent_1)
[agent_60]: 73 (agent_6)
```
After:
```
[agent_9]: 73
[agent_6]: 73
[agent_2]: 73
[agent_7]: 93
[agent_4]: 73
[agent_1]: 73
[agent_3]: 73
[agent_5]: 97
[agent_0]: 73
[agent_8]: 87
[agent_50]: 0
[agent_80]: 0
[agent_10]: 0
[agent_90]: 0
[agent_30]: 0
[agent_20]: 0
[agent_60]: 0
[agent_00]: 0
[agent_40]: 0
[agent_70]: 0
```
Closes#2948
Co-authored-by: Kevin Qian <kqian@google.com>
PiperOrigin-RevId: 826187198
Merge https://github.com/google/adk-python/pull/3219
## Summary
Enhance error messages for tool and agent not found errors to provide actionable guidance and reduce developer debugging time from hours to minutes.
Fixes#3217
## Changes
### Modified Files
1. **`src/google/adk/flows/llm_flows/functions.py`**
- Enhanced `_get_tool()` error message with:
- Available tools list (formatted, truncated to 20 for readability)
- Possible causes
- Suggested fixes
- Fuzzy matching suggestions
2. **`src/google/adk/agents/llm_agent.py`**
- Enhanced `__get_agent_to_run()` error message with:
- Available agents list (formatted, truncated to 20 for readability)
- Timing/ordering issue explanation
- Fuzzy matching for agent names
- Added `_get_available_agent_names()` helper method
### New Test Files
3. **`tests/unittests/flows/llm_flows/test_functions_error_messages.py`**
- Tests for enhanced tool not found error messages
- Fuzzy matching validation
- Edge cases (no close matches, empty tools dict, 100+ tools)
4. **`tests/unittests/agents/test_llm_agent_error_messages.py`**
- Tests for enhanced agent not found error messages
- Agent tree traversal validation
- Fuzzy matching for agents
- Long list truncation
## Testing Plan
### Unit Tests
```bash
pytest tests/unittests/flows/llm_flows/test_functions_error_messages.py -v
pytest tests/unittests/agents/test_llm_agent_error_messages.py -v
```
**Results**: âś… 8/8 tests passing
```
tests/unittests/flows/llm_flows/test_functions_error_messages.py::test_tool_not_found_enhanced_error PASSED
tests/unittests/flows/llm_flows/test_functions_error_messages.py::test_tool_not_found_fuzzy_matching PASSED
tests/unittests/flows/llm_flows/test_functions_error_messages.py::test_tool_not_found_no_fuzzy_match PASSED
tests/unittests/flows/llm_flows/test_functions_error_messages.py::test_tool_not_found_truncates_long_list PASSED
tests/unittests/agents/test_llm_agent_error_messages.py::test_agent_not_found_enhanced_error PASSED
tests/unittests/agents/test_llm_agent_error_messages.py::test_agent_not_found_fuzzy_matching PASSED
tests/unittests/agents/test_llm_agent_error_messages.py::test_agent_tree_traversal PASSED
tests/unittests/agents/test_llm_agent_error_messages.py::test_agent_not_found_truncates_long_list PASSED
8 passed, 1 warning in 4.38s
```
### Example Enhanced Error Messages
#### Before (Current Error)
```
ValueError: Function get_equipment_specs is not found in the tools_dict: dict_keys(['get_equipment_details', 'query_vendor_catalog', 'score_proposals'])
```
#### After (Enhanced Error)
```
Function 'get_equipment_specs' is not found in available tools.
Available tools: get_equipment_details, query_vendor_catalog, score_proposals
Possible causes:
1. LLM hallucinated the function name - review agent instruction clarity
2. Tool not registered - verify agent.tools list
3. Name mismatch - check for typos
Suggested fixes:
- Review agent instruction to ensure tool usage is clear
- Verify tool is included in agent.tools list
- Check for typos in function name
Did you mean one of these?
- get_equipment_details
```
## Community Impact
- **Addresses 3 active issues**: #2050, #2933 (12 comments), #2164
- **Reduces debugging time** from 3+ hours to < 5 minutes (validated in production multi-agent RFQ solution for recent partner nanothon initiative)
- **Improves developer experience** for new ADK users
## Implementation Details
- Uses standard library `difflib` for fuzzy matching (no new dependencies)
- Error path only (no performance impact on happy path)
- Measured performance: < 0.03ms per error
- Truncates long lists to first 20 items to prevent log overflow
- Fully backward compatible (same exception types)
## Checklist
- [x] Unit tests added and passing (8/8 tests)
- [x] Code formatted with `./autoformat.sh` (isort + pyink)
- [x] No new dependencies (uses standard library `difflib`)
- [x] Docstrings updated
- [x] Tested with Python 3.11
- [x] Issue #3217 created and linked
## Related Issues
- Fixes#3217
- Addresses #2050 - Tool verification callback request
- Addresses #2933 - How to handle "Function is not found in the tools_dict" Error
- Addresses #2164 - ValueError: {agent} not found in agent tree
---
**Note**: For production scenarios where LLM tool hallucinations occur, ADK's built-in [`ReflectAndRetryToolPlugin`](https://github.com/google/adk-python/blob/main/src/google/adk/plugins/reflect_retry_tool_plugin.py) can automatically retry failed tool calls (available since v1.16.0). This PR's enhanced error messages complement that by helping developers quickly identify and fix configuration issues during development.
Cheers, JP
Co-authored-by: Yvonne Yu <yyyu@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3219 from jpantsjoha:feat/better-error-messages a4df8bfb031685dce9e528d8eb7006f53447b75b
PiperOrigin-RevId: 826132579
Make sure _add_instructions_to_user_content skips over user messages that carry function_response parts so tool_use/tool_result blocks stay together
Close#3229
Co-authored-by: George Weale <gweale@google.com>
PiperOrigin-RevId: 826076141
Include a manual mode for testing first and will remove after verifying it works.
When a commit with `Merge https://github.com/google/adk-python/pull/3333` in description is pushed by copybara, the workflow will automatically close the PR.
Co-authored-by: Wei Sun (Jack) <weisun@google.com>
PiperOrigin-RevId: 826058313
Remove validation for output_schema and agent transfer flags.
The check that prevented `output_schema` from co-existing with agent transfer capabilities (`disallow_transfer_to_parent` or `disallow_transfer_to_peers` being False) has been removed. The agent will no longer automatically set these transfer flags to True when `output_schema` is present.
Co-authored-by: Ieva Grublyte <ievagrublyte@google.com>
PiperOrigin-RevId: 825998224
Merge https://github.com/google/adk-python/pull/3282
The `process_bind_param` and `process_result_value` methods in the `DynamicPickleType` class have been modified to handle MySQL dialect in addition to Spanner. This change ensures that pickled values are correctly processed for both database types.
**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: #3283
**2. Or, if no issue exists, describe the change:**
_If applicable, please follow the issue templates to provide as much detail as
possible._
**Problem:**
When using `DatabaseSessionService` with MySQL backend in google-adk v1.17.0, the application crashes with the following error: app.resources.runner:event_generator:260 - Error in event_generator: (builtins.TypeError) 'tuple' object cannot be interpreted as an integer
<img width="1237" height="129" alt="image" src="https://github.com/user-attachments/assets/0a5fc223-600a-4a92-8443-4d37fb1267f6" />
Root cause: The `DynamicPickleType` class in `database_session_service.py` configures MySQL dialect to use `LONGBLOB` for storing pickled data (line 117-118), but the `process_bind_param` and `process_result_value` methods only handle pickle serialization/deserialization for Spanner dialect, not MySQL. This causes MySQL to attempt storing raw Python objects instead of pickled bytes, leading to serialization errors and potential data corruption.
**Solution:**
Added MySQL to the pickle serialization logic in both `process_bind_param` and `process_result_value` methods, treating it the same way as Spanner dialect. This ensures that:
- Data is properly pickled to bytes before being stored in MySQL's LONGBLOB column
- Data is properly unpickled when retrieved from the database
- No breaking changes to existing functionality for other dialects (SQLite, PostgreSQL)
### 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:**
- [x] I have added or updated unit tests for my change.
- [x] All unit tests pass locally.
**Summary of `pytest` results:**
<img width="929" height="306" alt="image" src="https://github.com/user-attachments/assets/3d548b96-ac49-4101-8405-a289a722293c" />
**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
- [x] I have read the [CONTRIBUTING.md](https://github.com/google/adk-python/blob/main/CONTRIBUTING.md) document.
- [x] I have performed a self-review of my own code.
- [x] I have commented my code, particularly in hard-to-understand areas.
- [x] I have added tests that prove my fix is effective or that my feature works.
- [x] New and existing unit tests pass locally with my changes.
- [x] I have manually tested my changes end-to-end.
- [x] Any dependent changes have been merged and published in downstream modules.
### Additional context
_Add any other context or screenshots about the feature request here._
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3282 from hung12ct:fix/mysql-pickle-serialization d9df37adb7dfbbfd8502a0fe65c4f8bca3d0d978
PiperOrigin-RevId: 825834360
This CL introduces a new tool, get_job_info, to the BigQuery toolset. This tool allows retrieving metadata about a BigQuery job, such as slot usage, job configuration, statistics, and job status.
Closes#2928
Co-authored-by: Dongyu Jia <dongyuj@google.com>
PiperOrigin-RevId: 825762399
Session input file contains fields that are needed to run evals and later be able retrieve the session generated by them.
Co-authored-by: Ankur Sharma <ankusharma@google.com>
PiperOrigin-RevId: 825742522
We also change VertexAiSessionService and VertexAiMemoryBankService to both use keyword arguments for project, location, agent engine id, and express mode api key
PiperOrigin-RevId: 825719331
Merge https://github.com/google/adk-python/pull/3333
## Summary
Add ignore patterns for popular AI coding assistant configuration files to prevent committing developer-specific settings. This aligns with the project's approach of providing `AGENTS.md` as a general starting point that developers can symlink or copy and customize locally.
## Changes
Added `.gitignore` patterns for 10 popular AI coding tools:
- **Claude Code** - `.claude/`, `CLAUDE.md`
- **Cursor** - `.cursor/`, `.cursorrules`, `.cursorignore`
- **Windsurf** - `.windsurfrules`
- **Aider** - `.aider*`
- **Continue.dev** - `.continue/`
- **Codeium** - `.codeium/`
- **GitHub Next** - `.githubnext/`
- **Roo Code** - `.roo/`, `.rooignore`
- **Bolt** - `.bolt/`
- **v0** - `.v0/`
## Rationale
Each developer may want different AI tool configurations and personal instructions. By ignoring these files, we:
- Prevent accidental commits of personal AI assistant settings
- Keep the repository clean of developer-specific configurations
- Allow developers to customize their AI tools without affecting others
- Maintain consistency with the project's `AGENTS.md` approach
Co-authored-by: Yvonne Yu <yyyu@google.com>
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3333 from google:chore/ignore-ai-tool-configs 0abe4ccdf130ac93c3d2c353556f0ce7c305c429
PiperOrigin-RevId: 825682646
Usage: you can create symlink to AGENTS.md with the file name required by the specific AI coding tool.
Co-authored-by: Wei Sun (Jack) <weisun@google.com>
PiperOrigin-RevId: 825627185
Previously this will return true for events yielded from before_agent_callback when there are state changes.
Note with this change, it will also return false for state delta only callbacks even after main response, but this is fine as long as the actual final response event has it to be true.
Closes#2992
PiperOrigin-RevId: 825313208
Previously this will return true for events yielded from before_agent_callback when there are state changes.
Note with this change, it will also return false for state delta only callbacks even after main response, but this is fine as long as the actual final response event has it to be true.
Closes#2992
PiperOrigin-RevId: 825279439
Merge https://github.com/google/adk-python/pull/3037
fix: [#3036](https://github.com/google/adk-python/issues/3036)
- Fix FunctionTool parameter filtering to support CrewAI-style tools
- Functions with **kwargs now receive all parameters except 'self' and 'tool_context'
- Maintains backward compatibility with explicit parameter functions
- Add comprehensive tests for **kwargs functionality
Fixes parameter filtering issue where CrewAI tools using **kwargs pattern would receive empty parameter dictionaries, causing search_query and other parameters to be None.
#non-breaking
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3037 from omarcevi:fix/function-tool-kwargs-parameter-filtering 012bbfcfd68e83a29635ac74718a1bd1323c5187
PiperOrigin-RevId: 825275686
This change introduces a powerful pattern for customizing code execution environments by extending a base `CodeExecutor`. It showcases how to inject setup code to prepare the environment before a user's code is run, enabling advanced use cases that require specific configurations.
As a practical example, this change implements `CustomCodeExecutor`, a subclass of `VertexAiCodeExecutor`, to solve the problem of rendering non-standard characters in `matplotlib` plots (Issue #2993). The custom executor programmatically adds a Japanese font to the `matplotlib` font manager at runtime.
This is achieved by overriding the `execute_code` method to add font files to execution input and prepend the necessary font-loading logic. This approach is not limited to fonts and can be adapted for other setup tasks.
Fixes: #2993
PiperOrigin-RevId: 825240143
This change introduces a new sample agent and documentation to demonstrate the usage of the `VertexAiCodeExecutor`.
The new agent, located at `vertex_code_execution/agent.py`, is a direct counterpart to the existing sample at `code_execution/agent.py`. The key difference is that this new agent uses `VertexAiCodeExecutor` to execute code within the Vertex AI Code Interpreter Extension, whereas the original sample uses `BuiltInCodeExecutor` to run code in the model's built-in sandbox.
A `README.md` file is also included to provide an overview, setup instructions, and sample usage for the new agent.
Related: #2993
PiperOrigin-RevId: 825239758
This change introduces a new `detect_anomalies` tool in `query_tool.py` which uses BigQuery ML's `CREATE MODEL` with `ARIMA_PLUS` type and `ML.DETECT_ANOMALIES` to detect anomalies. The new function is also added to the `bigquery_toolset`.
PiperOrigin-RevId: 825181489
* feat: Added support for enums as arguments for function tools
* feat: Add default value support for function tools
fix: Add more test cases inside `test_build_function_declaration.py` for passing Enums as arguments
* fix: format code with pyink
---------
Co-authored-by: Wei Sun (Jack) <weisun@google.com>
Co-authored-by: Yvonne Yu <150068659+yyyu-google@users.noreply.github.com>
ADK already has a set of metrics that don't rely expected_invocations. Also, for eval cases with conversation scenario, this would be the main line case.
PiperOrigin-RevId: 825101481
Introduces the `BigQueryLoggingPlugin` for capturing and sending ADK lifecycle events to Google BigQuery. This allows for persistent storage and analysis of agent and tool interactions. The plugin supports asynchronous logging, automatic dataset/table creation, and comprehensive event capture.
Also refactors common formatting utilities (_format_content, _format_args) for shared use.
PiperOrigin-RevId: 824703739
This CL refactors VertexAiSessionService to use the asynchronous aio client for all Vertex AI API calls. This ensures that the service methods are non-blocking and can be used effectively in an asyncio environment.
PiperOrigin-RevId: 824573356
The computer_use sample now supports launching with a `user_data_dir` to maintain browser state across runs. The sample agent is updated to use a shared temporary directory for the browser profile, preserving login sessions and other data.
PiperOrigin-RevId: 823749082
While testing the bidi streaming sample agent, I noticed that the import was erroring and I think it's a typo -- the other bidi sample agents all import `from google.adk.agents.llm_agent`.
PiperOrigin-RevId: 823662586
Details:
- Adds the `LlmBackedUserSimulator` which uses an LLM to generate user prompts until it decides that the conversation is complete.
- Adds unit tests for the new functionality.
PiperOrigin-RevId: 823557910
This change adds a new section to the README.md, detailing past community events. The first entry is for the completed ADK's 1st community meeting, with links to the recording and deck.
PiperOrigin-RevId: 823322924
Right now the failure eats up the traceback information and there is no clear way for the developer to know what went wrong. Adding this traceback info could give them the needed debugging information.
PiperOrigin-RevId: 823179833
## What's Added
- **PostgreSQL MCP Agent** ([mcp_postgres_agent/agent.py](cci:7://file:///Users/admin/git%20repos/adk-python/contributing/samples/mcp_postgres_agent/agent.py:0:0-0:0)): A fully functional agent that connects to PostgreSQL databases via the `postgres-mcp` MCP server
- **Comprehensive README** ([mcp_postgres_agent/README.md](cci:7://file:///Users/admin/git%20repos/adk-python/contributing/samples/mcp_postgres_agent/README.md:0:0-0:0)): Documentation with setup instructions, configuration details, and example queries
- **Environment Configuration**: Support for secure credential management via `.env` files
## Key Features
- **MCP Integration**: Demonstrates proper use of `MCPToolset` with `StdioConnectionParams`
- **Zero Installation**: Uses `uvx` to run the MCP server without manual installation
- **Secure Credentials**: Database connection strings passed via environment variables
- **Production-Ready**: Uses Gemini 2.0 Flash with unrestricted access mode for full database operations
## Technical Details
- **Model**: Gemini 2.0 Flash
- **MCP Server**: `postgres-mcp` (via `uvx`)
- **Connection**: StdioConnectionParams with 60-second timeout
- **Environment Variable**: Maps `POSTGRES_CONNECTION_STRING` to `DATABASE_URI`
## Testing
The agent has been tested with:
- PostgreSQL database connections (local and remote)
- Schema inspection queries
- Data querying operations
- Table listing and management
## Example Queries
Users can interact with the agent using natural language queries like:
- "What tables are in the database?"
- "Show me the schema for the users table"
- "Query the first 10 rows from the products table"
This sample serves as a reference implementation for developers looking to integrate PostgreSQL databases with ADK agents using MCP.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3182 from Vrajesh-Babu:postgres-mcp f3b3846abae37ae376d3554624ac2b1be82f7adc
PiperOrigin-RevId: 822865931
The "What's new" section has been updated to highlight: Custom API Registration, Rewind functionality, and a new CodeExecutor utilizing the Vertex AI Code Execution Sandbox API. Previous updates on context compaction, resumability, ReflectRetryToolPlugin, and search tool support have been removed.
PiperOrigin-RevId: 822848112
The section detailing the first ADK community call scheduled for Oct 15, 2025, including date, time, meeting links, and agenda, has been removed.
The section detailing the first ADK community call scheduled for Oct 15, 2025, including date, time, meeting links, and agenda, has been removed.
PiperOrigin-RevId: 822843889
This release includes updates to the changelog with new features across Core, Evals, Integrations, Observability, Services, Tools, and UI, along with various bug fixes and improvements. The base CL number has also been updated.
PiperOrigin-RevId: 822730595
Since it appears in the same bubble as the rest of the LLM's response text, make it more human readable so it doesn't look out of place.
PiperOrigin-RevId: 822729061
The `BaseTool` expects the run_async to return a json-serializable object. By model_dump the McpTool result explicitly can allow what ADK runtime sees is identical to what is persisted in the session event list.
Before the change, runtime sees CallToolResult instance and Session persists its serialized dict.
https://github.com/modelcontextprotocol/python-sdk/blob/main/src/mcp/types.py#L916-L922
PiperOrigin-RevId: 822465432
From
```
You are an agent. Your internal name is "agent".
The description about you is "test description"
```
to
```
You are an agent. Your internal name is "agent". The description about you is "test description".
```
PiperOrigin-RevId: 822358196
To register a custom service:
- Create a factory function that takes a URI and returns an instance of your custom service. This function will parse any details it needs from the URI.
- Register your factory with the global service registry. You need to define a unique URI scheme for your service (e.g., custom).
PiperOrigin-RevId: 822310466
Merge https://github.com/google/adk-python/pull/3170
Addresses Feature Request: #3116
This PR adds a `speech_config` to the **LLM Agent configuration** for the **live use case**. When an **asynchronous LLM** call is made to the **Gemini Live API**, it prioritizes the most specific agent configuration's speech_config. If that is null, it then uses the run configuration's speech_config. Unit tests have been added to verify this behavior.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3170 from qyuo:bidi_agent_speech_config af1bd277d4f95c4a7d9aa0b16828ba3de826ce08
PiperOrigin-RevId: 822305427
Merge https://github.com/google/adk-python/pull/3194
Allow Google API toolsets to accept optional per-request headers
#3105
## Testing Plan
### Unit Tests
- âś… Added `test_init_with_additional_headers` in `test_google_api_tool.py` to verify headers are passed to RestApiTool
- âś… Added `test_prepare_request_params_merges_default_headers` in `test_rest_api_tool.py` to verify custom headers are merged into requests
- âś… Added `test_prepare_request_params_preserves_existing_headers` in `test_rest_api_tool.py` to verify critical headers (Content-Type, User-Agent) are not overridden by additional_headers
- âś… Updated `test_init` and `test_get_tools` in `test_google_api_toolset.py` to verify the parameter is properly stored and passed through
### Manual Testing
Tested with Google Ads API scenario (the original use case from issue #3105):
```python
import os
from google.adk.tools.google_api_tool import GoogleApiToolset
# Create toolset with developer-token header required by Google Ads API
google_ads_toolset = GoogleApiToolset(
client_id=os.environ["CLIENT_ID"],
client_secret=os.environ["CLIENT_SECRET"],
api_name="googleads",
api_version="v21",
additional_headers={"developer-token": os.environ["GOOGLE_ADS_DEV_TOKEN"]}
)
# Verify headers are included in API requests
tools = await google_ads_toolset.get_tools()
# Successfully made requests with the developer-token header
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3194 from Prhmma:feature/google-api-toolset-additional-headers-3105 e10489e82bfde5cf2bfd3f1bced3e1f5cea1f8b2
PiperOrigin-RevId: 822273582
Previously BuiltInCodeExecutor was missing the logic to save output files from executed code as artifacts, so images/visualizations wouldn't show up in the UI. This fix will iterate through all parts of the LlmResponse, and if any of them are images, it will save the image data using artifact_service (similar to what is done in VertexAICodeExecutor).
This fixes the backend, but there are still UI bugs that should be fixed -- events without content are currently ignored, so the image doesn't appear even though it is saved. We will add the UI fix in a separate change.
PiperOrigin-RevId: 822245140
- let _enforce_app_name_alignment warn instead of raising while caching the hint that now augments the existing “Session not found …” error
- tighten _infer_agent_origin so it ignores hidden folders (like .venv)
- make AgentTool reuse the parent runner’s app_name, stopping internal runners from conflicting in multi-agent setups
PiperOrigin-RevId: 822205860
Details:
- Adds the `StaticUserSimulator` which implements the current functionality of supplying a fixed set of user prompts for an EvalCase.
- Adds the `UserSimulatorProvider` which determines the type of user simulator required for an EvalCase (StaticUserSimulator or LlmBackedUserSimulator).
- Integrates the UserSimulatorProvider and UserSimulator into the CLI and evaluation infrastructure.
- Updates and adds unit tests for the new functionality.
- Miscellaneous updates to lay groundwork for a full implementation of the LlmBackedUserSimulator in the future.
PiperOrigin-RevId: 822198401
Merge https://github.com/google/adk-python/pull/3196
## Summary
Enhances the `AgentLoader` error message to provide clear guidance when users run `adk web` from incorrect directories.
## Motivation
During internal workshops, multiple teams encountered confusion when starting `adk web` from the wrong directory. This often happened when:
- Running `adk web my_agent/` instead of `adk web .`
- Being inside an agent directory when executing the command
- Configuring incorrect start paths during development
## Changes
- **Smart detection**: Checks if `agent.py` or `root_agent.yaml` exists in the current directory
- **Visual diagram**: Shows expected directory structure with actual agent name
- **Explicit command**: Includes `adk web <agents_dir>` usage example
- **Conditional hint**: Provides targeted guidance when user is detected to be inside an agent directory
## Example Error Message
### Before
```
ValueError: No root_agent found for 'my_agent'. Searched in 'my_agent.agent.root_agent', 'my_agent.root_agent' and 'my_agent/root_agent.yaml'. Ensure 'path/my_agent' is structured correctly, an .env file can be loaded if present, and a root_agent is exposed.
```
### After
```
ValueError: No root_agent found for 'my_agent'. Searched in 'my_agent.agent.root_agent', 'my_agent.root_agent' and 'my_agent/root_agent.yaml'.
Expected directory structure:
<agents_dir>/
my_agent/
agent.py (with root_agent) OR
root_agent.yaml
Then run: adk web <agents_dir>
Ensure 'path/my_agent' is structured correctly, an .env file can be loaded if present, and a root_agent is exposed.
HINT: It looks like you might be running 'adk web' from inside an agent directory. Try running 'adk web .' from the parent directory that contains your agent folder, not from within the agent folder itself.
```
## Testing
- âś… Existing unit tests pass (17/22, with 5 pre-existing failures unrelated to this change)
- âś… `test_agent_not_found_error` passes, confirming error message enhancement works correctly
- âś… Code follows ADK contribution guidelines
## Type
- [x] Bug fix (improved error messaging)
- [ ] Feature
- [ ] Breaking change
- [ ] Documentation
## Related
Fixes#3195
---
**Tags**: #non-breaking
🤖 Generated with [Claude Code](https://claude.com/claude-code)
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3196 from jpantsjoha:fix/improve-adk-web-error-message a73b190f5b021dbe0afa8426172696ee9eeae8da
PiperOrigin-RevId: 822186700
Merge https://github.com/google/adk-python/pull/3060
## Description
Fixes#3059
This PR fixes two endpoints in `adk web` that fail when using App objects instead of bare agents.
## Changes
- **Eval execution endpoint** (line ~969): Extract root_agent from App objects before passing to LocalEvalService
- **Graph visualization endpoint** (line ~1308): Extract root_agent from App objects before graph operations
Both endpoints now properly handle both BaseAgent and App objects by checking the type and extracting `.root_agent` when needed.
## Testing Plan
### Manual E2E Testing with ADK Web
Tested with an App object that includes context caching:
```python
from google.adk.apps import App
from google.adk.agents import LlmAgent
root_agent = LlmAgent(name="MyAgent", model="gemini-1.5-pro-002")
app = App(
name="my_agent",
root_agent=root_agent,
context_cache_config=ContextCacheConfig(...)
)
```
**Before fix:**
- Graph visualization failed (tried to call agent methods on App object)
- Eval execution failed (LocalEvalService received App instead of agent)
**After fix:**
- Graph visualization works correctly
- Eval execution works correctly
- Both endpoints properly extract root_agent from App objects
## Checklist
- [x] Code follows project style (autoformat.sh passed)
- [x] Changes are focused and minimal
- [x] Issue #3059 created and referenced
- [x] Manual E2E testing completed
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/3060 from ejfn:ejfn/bugfix-app-object-endpoints 01c30191bfd9487a8c8463ccf24b297cb9a4ce37
PiperOrigin-RevId: 821746910
Add a header_provider param which is a callable[ReadonlyContext, Dict[str, Any]] for users to build headers in MCPToolset
fix: https://github.com/google/adk-python/issues/3156
PiperOrigin-RevId: 820412372
This change introduces type aliases for request and event conversion functions:
- `A2ARequestToADKRunArgsConverter`: For converting A2A `RequestContext` to an `ADKRunArgs` Pydantic model.
- `AdkEventToA2AEventsConverter`: For converting ADK `Event` to a list of A2A `A2AEvent` objects.
The `convert_a2a_request_to_adk_run_args` function now returns a structured `ADKRunArgs` model instead of a generic dictionary, improving type safety.
These converter types can now be provided via the `A2aAgentExecutorConfig` to customize the conversion logic used by the `A2aAgentExecutor`. The executor defaults to the existing `convert_a2a_request_to_adk_run_args` and `convert_event_to_a2a_events` functions if no custom converters are specified.
This allows users to inject their own logic for handling request and event conversions, for example, to add custom metadata or transform data types, without modifying the core executor.
PiperOrigin-RevId: 819934960
Merge https://github.com/google/adk-python/pull/2884closes: #2883
# Fix
When put leage data into event and load it. the _pickle.UnpicklingError was occurred.
The root caurse is `DynamicPickleType` mapping `BLOB` as default in case of MySql, not `LONGBLOB`. And learge data will be able to cut off tail of data. And raise pickle error.
# What todo
Defined `LONFBLOB` as default explicitly.
# Question
Where should we code the test code like this case? I cannot found the test code the DB and table was created expectedly.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2884 from Lin-Nikaido:fix/#2883-mysql-datatype-fix 2be9b38fc3f5d5083b0b6715a2bf7b4eff5d947b
PiperOrigin-RevId: 819891727
Merge https://github.com/google/adk-python/pull/2206
### Summary
This PR adds support for `ContextWindowCompressionConfig` in `RunConfig`.
This enables context window compression using a `trigger_tokens` threshold and a sliding window with a `target_tokens` limit.
This feature is useful for managing long-running audio inputs.
### Related Issue
Closes#2188
### Testing Plan
- Added new unit test: `test_streaming_with_context_window_compression_config`
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2206 from ac-machache:support/add-context-compression-config c8a5b15cae2d2b72f331797d07ae0bbaf977ed3c
PiperOrigin-RevId: 819855786
This change removes the `convert_session_to_eval_format` function and its associated unit tests. New tests for `create_gcs_eval_managers_from_uri` are also added.
PiperOrigin-RevId: 819576620
- add a shared --structured_logs flag to adk web and adk api_server so users can opt into JSON-formatted output
- introduce CloudTraceJSONFormatter that emits structured entries and attaches current Cloud Trace/Span IDs when an OpenTelemetry context is active
- update CLI logging setup to clear duplicate stdout handlers when Cloud Logging is enabled and to reconfigure existing handlers (like from Uvicorn) so they also pick up the structured format and requested log level
With the flag disabled the CLIs keep their existing text logs; when enabled, the services now produce Cloud Logging–friendly JSON that can be correlated with distributed traces.
PiperOrigin-RevId: 818823818
Update plugin manager and built-in plugins to prioritize CallbackContext. Keep InvocationContext access for legacy plugins with adapter. Change callback docs/tests to cover the new context.
PiperOrigin-RevId: 818822267
Update plugin manager and built-in plugins to prioritize CallbackContext. Keep InvocationContext access for legacy plugins with adapter. Change callback docs/tests to cover the new context.
PiperOrigin-RevId: 818798087
This is so we don't need to worry about side effect of Loop in all agent type. Custom agent should do the same if there exists loop inside.
PiperOrigin-RevId: 818766305
This change removes the `run_evals` function and its helper `_get_evaluator` from `cli_eval.py`, as they were marked as deprecated. Corresponding test mocks and patches in `test_fast_api.py` are also removed.
PiperOrigin-RevId: 818719422
changed the LiteLLM content conversion so Part.file_data.file_uri (like the gs://…) becomes a file object with file_id, making sure GCS-backed files reach LiteLLM proxies instead of being dropped add unit tests covering both _get_content and _content_to_message_param paths for file URIs
PiperOrigin-RevId: 817658432
This change removes the `evaluate`, `_evaluate_row`, `are_tools_equal`, `_remove_tool_outputs`, `_report_failures`, and `_print_results` static methods from `TrajectoryEvaluator`, along with their corresponding unit tests. These methods were previously marked as deprecated.
PiperOrigin-RevId: 817477494
This CL updates the "What's new" section to include Resumability, ReflectRetryToolPlugin, Context compaction, and Search tool support. It also moves "Agent Config" and "Tool Confirmation" from "What's new" to "Key Features".
PiperOrigin-RevId: 817469210
The added section provides details for the community call on Oct 15, 2025, including the agenda and links to join and add to calendars.
PiperOrigin-RevId: 817457276
Agent developers can now create an eval set and add eval cases through command line itself. Adding an eval case is limited only to specifying conversation scenarios.
Sample comamnds:
- Create an eval set:
adk eval_set create \
contributing/samples/hello_world \
set_01
- Add an eval case with scenario file
Content of scenarios.json file:
'{"scenarios": [{"starting_prompt": "hello", "conversation_plan": "world"}]}'
adk eval_set add_eval_case \
contributing/samples/hello_world \
set_01 \
--scenarios scenarios.json
PiperOrigin-RevId: 817456117
The `agent_loader.load_agent` method can now return an `App` object. This change unwraps the `App` to get its `root_agent` before passing it to the graph builder, makes sure a `BaseAgent` instance is always used
PiperOrigin-RevId: 817209601
Details:
- Introduces a concept of `ConversationScenario` to represent a scenario that user simulator is supposed to follow.
- Introduces a `UserSimulator` interface, that one should implement. UserSimulator interface will be integrated with LocalEvalService in subsequent PRs.
PiperOrigin-RevId: 816883699
When there are multiple intervals and compactions, the original implementation only keep the last one. The right implementation is to keep as many compaction events/summary as the requested internals.
PiperOrigin-RevId: 816516662
This plugin intercepts tool failures, provides structured guidance to the LLM for reflection and correction, and retries the operation up to a configurable limit.
**Key Features:**
- **Concurrency Safe:** Uses locking to safely handle parallel tool
executions
- **Configurable Scope:** Tracks failures per-invocation (default) or globally
using the `TrackingScope` enum.
- **Extensible Scoping:** The `_get_scope_key` method can be overridden to
implement custom tracking logic (e.g., per-user or per-session).
- **Granular Tracking:** Failure counts are tracked per-tool within the
defined scope. A success with one tool resets its counter without affecting
others.
- **Custom Error Extraction:** Supports detecting errors in normal tool
responses
that
don't throw exceptions, by overriding the `extract_error_from_result`
method.
**Example:**
```python
from my_project.plugins import ReflectAndRetryToolPlugin, TrackingScope
# Example 1: (MOST COMMON USAGE):
# Track failures only within the current agent invocation (default).
error_handling_plugin = ReflectAndRetryToolPlugin(max_retries=3)
# Example 2:
# Track failures globally across all turns and users.
global_error_handling_plugin = ReflectAndRetryToolPlugin(max_retries=5,
scope=TrackingScope.GLOBAL)
# Example 3:
# Retry on failures but do not throw exceptions.
error_handling_plugin =
ReflectAndRetryToolPlugin(max_retries=3,
throw_exception_if_retry_exceeded=False)
# Example 4:
# Track failures in successful tool responses that contain errors.
class CustomRetryPlugin(ReflectAndRetryToolPlugin):
async def extract_error_from_result(self, *, tool, tool_args,tool_context,
result):
# Detect error based on response content
if result.get('status') == 'error':
return result
return None # No error detected
error_handling_plugin = CustomRetryPlugin(max_retries=5)
```
PiperOrigin-RevId: 816456549
Merge https://github.com/google/adk-python/pull/2857
Adds support for invoking Gemma models via the Gemini API endpoint. To support agentic function, callbacks are added which can extract and transform function calls and responses into user and model messages in the history.
This change is intended to allow developers to explore the use of Gemma models for agentic purposes without requiring local deployment of the models. This should ease the burden of experimentation and testing for developers.
A basic "hello world" style agent example is provided to demonstrate proper functioning of Gemma 3 models inside an Agent container, using the dice roll + prime check framework of similar examples for other models.
## Testing
### Testing Plan
- add and run integration and unit tests
- manual run of example `multi_tool_agent` from quickstart using new `Gemma` model
- manual run of `hello_world_gemma` agent
### Automated Test Results:
| Test Command | Results |
|----------------|---------|
| pytest ./tests/unittests | 4386 passed, 2849 warnings in 58.43s |
| pytest ./tests/unittests/models/test_google_llm.py | 100 passed, 4 warnings in 5.83s |
| pytest ./tests/integration/models/test_google_llm.py | 5 passed, 2 warnings in 3.73s |
### Manual Testing
Here is a log of `multi_tool_agent` run with locally-built wheel and using Gemma model.
```
❯ adk run multi_tool_agent
Log setup complete: /var/folders/bg/_133c0ds2kb7cn699cpmmh_h0061bp/T/agents_log/agent.20250904_152617.log
To access latest log: tail -F /var/folders/bg/_133c0ds2kb7cn699cpmmh_h0061bp/T/agents_log/agent.latest.log
/Users/<redacted>/venvs/adk-quickstart/lib/python3.11/site-packages/google/adk/cli/cli.py:143: UserWarning: [EXPERIMENTAL] InMemoryCredentialService: This feature is experimental and may change or be removed in future versions without notice. It may introduce breaking changes at any time.
credential_service = InMemoryCredentialService()
/Users/<redacted>/venvs/adk-quickstart/lib/python3.11/site-packages/google/adk/auth/credential_service/in_memory_credential_service.py:33: UserWarning: [EXPERIMENTAL] BaseCredentialService: This feature is experimental and may change or be removed in future versions without notice. It may introduce breaking changes at any time.
super().__init__()
Running agent weather_time_agent, type exit to exit.
[user]: what's the weather like today?
[weather_time_agent]: Which city are you asking about?
[user]: new york
[weather_time_agent]: OK. The weather in New York is sunny with a temperature of 25 degrees Celsius (77 degrees Fahrenheit).
```
And here is a snippet of a log generated with DEBUG level logging of the `hello_world_gemma` sample. It demonstrates how function calls are extracted and inserted based on Gemma model interactions:
```
...
2025-09-04 15:32:41,708 - DEBUG - google_llm.py:138 -
LLM Request:
-----------------------------------------------------------
System Instruction:
None
-----------------------------------------------------------
Contents:
{"parts":[{"text":"\n You roll dice and answer questions about the outcome of the dice rolls.\n You can roll dice of different sizes...\n"}],"role":"user"}
{"parts":[{"text":"Hi, introduce yourself."}],"role":"user"}
{"parts":[{"text":"Hello! I am data_processing_agent, a hello world agent that can roll many-sided dice and check if numbers are prime. I'm ready to assist you with those tasks. Let's begin!\n\n\n\n"}],"role":"model"}
{"parts":[{"text":"Roll a die with 100 sides and check if it is prime"}],"role":"user"}
{"parts":[{"text":"{\"args\":{\"sides\":100},\"name\":\"roll_die\"}"}],"role":"model"}
{"parts":[{"text":"Invoking tool `roll_die` produced: `{\"result\": 82}`."}],"role":"user"}
{"parts":[{"text":"{\"args\":{\"nums\":[82]},\"name\":\"check_prime\"}"}],"role":"model"}
{"parts":[{"text":"Invoking tool `check_prime` produced: `{\"result\": \"No prime numbers found.\"}`."}],"role":"user"}
{"parts":[{"text":"The die roll was 82, and it is not a prime number.\n\n\n\n"}],"role":"model"}
{"parts":[{"text":"Roll it again."}],"role":"user"}
-----------------------------------------------------------
Functions:
-----------------------------------------------------------
2025-09-04 15:32:41,708 - INFO - models.py:8165 - AFC is enabled with max remote calls: 10.
2025-09-04 15:32:42,693 - INFO - google_llm.py:180 - Response received from the model.
2025-09-04 15:32:42,693 - DEBUG - google_llm.py:181 -
LLM Response:
-----------------------------------------------------------
Text:
{"args":{"sides":100},"name":"roll_die"}
-----------------------------------------------------------
...
```
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2857 from douglas-reid:add-gemma-via-api e6d015f6a9ccbcf20ef7a7af8e4bbe1e9a5936b6
PiperOrigin-RevId: 816451001
If `EventsCompactionConfig` is provided without a `compactor`, a `SlidingWindowCompactor` is now automatically instantiated using the `root_agent`'s LLM. This simplifies configuration by providing a sensible default.
PiperOrigin-RevId: 816038579
The class is now named `LlmEventSummarizer` to better reflect that its primary function is to use an LLM to summarize events. The docstring has been updated to clarify that this class is responsible *only* for the LLM-based summarization of a given set of events, while the logic for determining *when* and *which* events form the sliding window is handled by an external component, such as an ADK Runner.
PiperOrigin-RevId: 815976264
This change introduces a new `analyze_contribution` function in `query_tool.py` which uses BigQuery ML's `CREATE MODEL` with `CONTRIBUTION_ANALYSIS` type and `ML.GET_INSIGHTS` to analyze the contribution of different dimensions to a given metric. The new function is also added to the `bigquery_toolset`.
PiperOrigin-RevId: 815849281
before this change, we estimate the token count of the contents to cache and use it to compare with the threshold user set. but that's not precise , so we use the actual prompt token count of previous llm request.
We won't create cache for the very initial request
PiperOrigin-RevId: 814484840
We updated the one of the public methods on AgentEvaluator to take in eval metric configurations using a more formal EvalConfig data model.
We also mark "criteria" field on the method as deprecated.
Updated some integration test cases.
PiperOrigin-RevId: 814314134
The root cause is an unsafe in-memory mutation. The `SaveFilesAsArtifactsPlugin` was saving a direct reference to the message part and then modifying the message list in-place. This created a race condition where downstream code could alter the original part *after* it had been saved as an artifact, leading to a corrupted state.
This CL saves a `copy.copy()` of the artifact, which create a snapshot of the data.
Also Changes the plugin to return a new `types.Content` object instead of modifying the original message in-place
PiperOrigin-RevId: 814308070
This is allow user to update session state without running the agent. e.g. if I want to test some case when session has certain state on adk web.
PiperOrigin-RevId: 814252851
Currently, the A2A Task -> ADK event conversion is producing the same events on the last two update events (the last is a status update marking the task complete)
The change here based on A2AClientEvent(task, update):
- if the update == None: handle the non-streaming task case and also streaming case for the initial task creation event
- if the update = TaskStatusUpdateEvent AND a message is set: emit an event with that message
- if a task status update AND no message is set: don't emit event (for example, the final status update)
- if the update is ArtifactUpdateEvent and it's final artifact: emit the event
PiperOrigin-RevId: 812878869
The PR does two main things:
1) Introduces a new rubric based tool use metric
2) Given that we now have two rubric based metric, we refactor and create a new RubricBasedEvaluator interface.
PiperOrigin-RevId: 811983514
mainly because http://github.com/robots.txt disallows `/*/raw/` path. using GCS HTTP URIs is more reliable with Gemini model.
PiperOrigin-RevId: 811409688
Changes include:
- Implementing missing attributes. e.g. 'gen_ai.agent.name'
- Specifying reasons for not filling out some conditionally required attributes. e.g. 'gen_ai.data_source.id'
- Specifying reasons for not including certain attributes which are specified in current semconv. e.g. inference attributes on agent spans
PiperOrigin-RevId: 811379706
this is to allow turning on debug log for debugging if some unexpected behavior observed during running cache analysis experiments.
PiperOrigin-RevId: 811189954
This is to avoid serialization issue for some fields that are not json serializable.
meanwhile restructure the debug logs in context cache manager for better debugging potential issues.
PiperOrigin-RevId: 811182492
AppDetails require two pieces of information:
1) Instructions
2) Tools
Both these pieces of information are gathered using the llm_request that was passed to the model. This approach, slightly invasive, ensures that we capture the "exact" instructions and tools that were given to the model.
PiperOrigin-RevId: 811180648
Details:
1. Data model for storing App Details (the agentic system)
As we move towards LLM as Judge metrics, we see that some of these metrics need information about the Agentic system that was used for inferencing. We add a data model to capture that.
2. Data model for Steps
We refine the concept of intermediate data. Previously it stored data in the form of a multiple lists, thereby losing out on the chronological information. This information is needed for some of the metrics. So we refine the concept of intermediate data as series of logical steps that an Agent Take.
PiperOrigin-RevId: 811122784
Merge https://github.com/google/adk-python/pull/2823
Description
This change introduces a tool_name_prefix attribute to McpToolset and McpToolsetConfig. This allows for adding a prefix to the
names of all tools within the toolset, which can help avoid naming collisions and provide better organization.
The implementation involves updating the McpToolset's __init__ and from_config methods to handle the new tool_name_prefix and
adding the corresponding field to McpToolsetConfig.
Testing Plan
A new unit test file has been added to ensure the functionality works as expected.
- `tests/unittests/tools/test_mcp_toolset.py`:
- The test_mcp_toolset_with_prefix test case verifies that the tool_name_prefix is correctly applied to the tool names
retrieved from the toolset.
- All tests were run via pytest and passed.
Related Issue
- Closes#2814
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2823 from shsha4:fix/issue-2814 e8e5b0d6d5f406d3875faf2229a96701725b7a5e
PiperOrigin-RevId: 810500616
Merge https://github.com/google/adk-python/pull/2458
**Summary**
Verifies that user-provided messages are always passed to the LLM as 'user' role, regardless of whether the role is explicitly set in types.Content. Before the current fix, if the LlmRequest from the user doesn't have the 'user' role, but has the user content, then the text is being replaced with the standard text - "Handle the requests as specified in the System Instruction." and the content from the user is completely ignored and not passed into the LLM.
**Code to replicate the problem**
```
from google.adk.agents import LlmAgent
from google.adk.sessions import InMemorySessionService
from google.adk.runners import Runner
from google.genai.types import Content, Part
from google.adk.models.lite_llm import LiteLlm
from google.adk.models import LlmRequest
from google.genai import types
from pydantic import Field
import litellm
litellm._turn_on_debug()
import warnings
warnings.filterwarnings("ignore", category=UserWarning, message=".*InMemoryCredentialService.*")
import os
from dotenv import load_dotenv
# Load environment variables from the agent directory's .env file
load_dotenv()
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
# Define agent with output_key
root_agent = LlmAgent(
name="name_of_agent",
model=LiteLlm(model="azure/gpt-4o-mini"),
instruction="You are a customer agent to help the users with their concerns."
)
# --- Setup Runner and Session ---
app_name, user_id, session_id = "state_app", "user1", "session1"
session_service = InMemorySessionService()
runner = Runner(
agent=root_agent,
app_name=app_name,
session_service=session_service
)
print(f"Runner created for agent '{runner.agent.name}'.")
session = await session_service.create_session(
app_name=app_name,
user_id=user_id,
session_id=session_id
)
# --- Run the Agent ---
async def call_agent_async(query: str, runner, user_id, session_id):
user_message = Content(parts=[Part(text=query)])
async for event in runner.run_async(
user_id=user_id,
session_id=session_id,
new_message=user_message
):
print("event")
print(f" [Event]\n Author: {event.author}\n Type: {type(event).__name__}",
f"\n Final: {event.is_final_response()}\n Content: {event.content}")
return event
event = await call_agent_async("What is the capital of India.",runner=runner,user_id=user_id,session_id=session_id)
```
**Before the fix (current adk-python code output)**
```
00:29:24 - LiteLLM:DEBUG: utils.py:348 -
00:29:24 - LiteLLM:DEBUG: utils.py:348 - Request to litellm:
00:29:24 - LiteLLM:DEBUG: utils.py:348 - litellm.acompletion(model='azure/gpt-4o-mini', messages=[{'role': 'developer', 'content': 'You are a customer agent to help the users with their concerns.\n\nYou are an agent. Your internal name is "name_of_agent".'}, {'role': 'user', 'content': 'Handle the requests as specified in the System Instruction.'}], tools=None, response_format=None)
```
**After the fix (after resolving the fix)**
```
00:28:46 - LiteLLM:DEBUG: utils.py:349 -
00:28:46 - LiteLLM:DEBUG: utils.py:349 - Request to litellm:
00:28:46 - LiteLLM:DEBUG: utils.py:349 - litellm.acompletion(model='azure/gpt-4o-mini', messages=[{'role': 'developer', 'content': 'You are a customer agent to help the users with their concerns.\n\nYou are an agent. Your internal name is "name_of_agent".'}, {'role': 'user', 'content': 'What is the capital of India.'}], tools=None, response_format=None)
```
**Testing**
Following unit test is created to test the applied changes and added in the location as suggested in the guidelines.
adk-python\tests\unittests\models\test_base_llm.py
```
import pytest
from google.genai import types
from google.adk.models.llm_request import LlmRequest
from google.adk.models.lite_llm import _get_completion_inputs
@pytest.mark.parametrize("content_kwargs", [
# Case 1: Explicit role provided
{"role": "user", "parts": [types.Part(text="This is an input text from user.")]},
# Case 2: Role omitted, should still be treated as 'user'
{"parts": [types.Part(text="This is an input text from user.")]}
])
def test_user_content_role_defaults_to_user(content_kwargs):
"""
Verifies that user-provided messages are always passed to the LLM as 'user' role,
regardless of whether the role is explicitly set in types.Content.
The helper `_get_completion_inputs` should give normalize messages so that
explicit 'user' and implicit (missing role) are equivalent.
"""
llm_request = LlmRequest(
contents=[types.Content(**content_kwargs)],
config=types.GenerateContentConfig()
)
messages, _, _, _ = _get_completion_inputs(llm_request)
assert all(
msg.get("role") == "user" for msg in messages
), f"Expected role 'user' but got {messages}"
assert any(
"This is an input text from user." == (msg.get("content") or "")
for msg in messages
), f"Expected the user text to be preserved, but got {messages}"
```
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2458 from TanejaAnkisetty:bug/agent-user-content 381b01418d249b9e6bd91ebb518ff25339a8e47b
PiperOrigin-RevId: 809281620
Static instructions:
Always added to system instructions for context caching
Dynamic instructions:
Added to system instructions when no static instruction exists (for backward compatibility), OR inserted before last batch of continuous user content when static instructions exist
PiperOrigin-RevId: 809170679
1. add a context cache config in app level which will apply to all agents in the app
2. pass on cache config through invocation context to llm_reqeust
3. store cache metadata in llm_response
4. lookup old cache metadata from latest event for reusing old cache
5. create new cache if old cache cannot be reused
PiperOrigin-RevId: 809158578
Currently there is chance for Cloud Monitoring-related errors in logs during shutdown. Let's disable metrics part until it is fixed.
PiperOrigin-RevId: 808930635
The docstrings for `compaction_range` and `compacted_content` are updated to reflect that compaction is based on timestamp ranges rather than sequence IDs, and to use consistent terminology ("compacted" instead of "summarized").
PiperOrigin-RevId: 808770610
Merge https://github.com/google/adk-python/pull/2960
1. All in one authentication sample (has an IDP, Agent and the application) under `contributing/samples/authn-adk-all-in-one/`
2. Documented for all the steps.
3. OAuth 2.0 Authorization Code Grant type used by the agent.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2960 from nikhilpurwant:main dfcc821602d265c4ae7cc42eb1f5739beaad6f87
PiperOrigin-RevId: 808672120
This add `GoogleMapsGroundingTool`, a built-in tool for Gemini 2 models to ground query results with Google Maps. This tool operates internally within the model and is only available when using the VertexAI Gemini API.
PiperOrigin-RevId: 808650501
Provide a more efficient way to compact LLM context for better agentic performance.
* `app`: the top level abstraction for an ADK application. It contains an root agent, and plugins.
* `content_strategy`: the abstraction for selecting the contents for LLM request.
* `compaction_strategy`: the abstraction for compacting the events.
* Added `sequence_id` and `summary_range` in event class.
PiperOrigin-RevId: 808634224
Merge https://github.com/google/adk-python/pull/2937
**Closes #2936**
This Pull Request addresses the issue where `LlmAgent` outputs, when configured with `output_schema` and `tools`, were presenting escaped Latin characters (e.g., `\xf3` for `Ăł`) in the final response. This behavior occurred because `json.dumps` was being called with `ensure_ascii=True` (its default), which is not ideal for human-readable output, especially when dealing with non-ASCII characters common in many languages like Portuguese.
**Changes Proposed:**
* Modified the `_OutputSchemaRequestProcessor` in `src/google/adk/flows/llm_flows/_output_schema_processor.py` to explicitly set `ensure_ascii=False` when calling `json.dumps` for the `set_model_response` tool's output.
**Impact:**
This change ensures that all non-ASCII characters in the structured model response are preserved in their natural form, improving the readability and user experience of agent outputs, particularly for users interacting in languages with accented characters or other special symbols.
**Testing:**
The fix was verified locally by running an `LlmAgent` with an `output_schema` and confirming that responses containing Latin characters (e.g., "ação", "caminhĂŁo", "Ăcone") are now correctly displayed without escaping.
COPYBARA_INTEGRATE_REVIEW=https://github.com/google/adk-python/pull/2937 from amenegola:fix/issue-2936-escape-chars 6cac00f97aa4cd8d8ccaa97ec5fffc74f57995dc
PiperOrigin-RevId: 808622892
- Add conformance command group with create subcommand
- Implement category/name/spec.yaml with generated-*.yaml files
- Support executing agents with queries and recording sessions
- Create test cases with recorded llm interactions and tool calls/results
Expected folder structure:
```
conformance_repo/
├── agents/ # Agent definitions - contains all config-based agents shared by test cases.
│ ├── single_basic/
│ ├── multi_basic/
│ └── single_tool_builtin/
│
└── tests/ # Test cases
├── core/ # Test category
│ ├── desc_001/ # Individual test case
│ │ ├── spec.yaml # Human-written specification
│ │ ├── generated-session.yaml
│ │ ├── generated-recordings.yaml
│ │ └── ... # Potential future generated files
│ ├── f_001/
│ │ ├── spec.yaml
│ │ ├── generated-session.yaml
│ │ ├── generated-recordings.yaml
│ │ └── ...
```
Help text:
```
-> % adk conformance create --help
Usage: adk conformance create [OPTIONS] [PATHS]...
Generate ADK conformance test YAML files from TestCaseInput specifications.
NOTE: this is work in progress.
This command reads TestCaseInput specifications from input.yaml files, executes the specified test cases against agents, and generates conformance test files with recorded agent interactions as
test.yaml files.
Expected directory structure: category/name/input.yaml (TestCaseInput) -> category/name/test.yaml (TestCase)
PATHS: One or more directories containing test case specifications. If no paths are provided, defaults to 'tests/' directory.
Examples:
Use default directory: adk conformance create
Custom directories: adk conformance create tests/core tests/tools
Options:
--help Show this message and exit.
```
PiperOrigin-RevId: 808609547
Corrected `CountInvocationPlugin` to be a class reference and added `ContextFilterPlugin` to limit the number of tool invocations kept in the context to 3.
PiperOrigin-RevId: 808591608
When start the server with `--extra_plugins=google.adk.cli.plugins.recordings_plugin.RecordingsPlugin`, it will trigger recording with expected state in session.
PiperOrigin-RevId: 808432022
This commit introduces a new ContextFilterPlugin which allows for filtering the LlmRequest contents before they are sent to the LLM. This helps in managing and potentially reducing the size of the LLM context.
The plugin provides two primary filtering mechanisms:
num_invocations_to_keep: Keeps only the specified number of the most recent user-model invocations. An invocation is defined as one or more user messages followed by a model response.
custom_filter: Allows for a user-defined callable to be applied to the contents for more flexible filtering.
Unit tests have been added to cover the different filtering scenarios, including:
Filtering by the last N invocations.
Filtering using a custom function.
Combining both filtering methods.
Handling cases with multiple user turns in a single invocation.
Ensuring no filtering occurs when options are not provided.
Gracefully handling exceptions from custom filter functions."
For example, when num_of_innovacations=2:
-----------------------------------------------------------
Contents:
{"parts":[{"text":"9"}],"role":"user"}
{"parts":[{"text":"I am sorry, I cannot fulfill this request. I need more information on what you would like me to do. I can roll a die or check prime numbers.\n"}],"role":"model"}
{"parts":[{"text":"1"}],"role":"user"}
{"parts":[{"text":"I am sorry, I cannot fulfill this request. I need more information on what you would like me to do. I can roll a die or check prime numbers.\n"}],"role":"model"}
{"parts":[{"text":"10"}],"role":"user"}
-----------------------------------------------------------
PiperOrigin-RevId: 808355316
Right now the bigquery sample agent is configured to run with OAuth, which requires some set up. This change makes it more readily usable, both locally and in AgentEngine, as Application Default Credentials (ADC) is easier to set up, and often local and AgentEngine environment already have it set up.
PiperOrigin-RevId: 808315879
Also moves the `Recordings` pydantic models into this plugins/ package.
Key features:
- Records LLM requests/responses and tool calls/results to YAML files in `generated-recordings.yaml`.
- Use session state to determine where to read and output recordings.
PiperOrigin-RevId: 807969100
Cloud Trace, Cloud Monitoring and Cloud Logging integrations are set up via OTel if otel_to_cloud CLI param/fast_api arg is provided.
This is similar to current Cloud Trace integration via trace_to_cloud, just extended to Monitoring and Logging as well.
PiperOrigin-RevId: 807385680
Cloud Trace, Cloud Monitoring and Cloud Logging integrations are set up via OTel if otel_to_cloud CLI param/fast_api arg is provided.
This is similar to current Cloud Trace integration via trace_to_cloud, just extended to Monitoring and Logging as well.
PiperOrigin-RevId: 807285744
Cloud Trace, Cloud Monitoring and Cloud Logging integrations are set up via OTel if otel_to_cloud CLI param/fast_api arg is provided.
This is similar to current Cloud Trace integration via trace_to_cloud, just extended to Monitoring and Logging as well.
PiperOrigin-RevId: 807230668
The `after_agent_callback` in plugin works similarly as the `after_agent_callback` in `base_agent.py`, e.g. it only append new content, but cannot modify the previous content.
PiperOrigin-RevId: 807162139
Similarity search tool supports similarity search on Spanner data by embedding a text query to a vector and run vector search with the embedded vector.
PiperOrigin-RevId: 806502499
Recent change to the updated A2A Client SDK broke the logging utilities. This updates those logging utilities to work with the new A2A SDK structure.
PiperOrigin-RevId: 806482017
Right now the tolls are always running against multi-region US by default. With this change the agent builder can scope the tools to data and compute in a particular BigQuery location.
PiperOrigin-RevId: 806473857
Update the bug report issue template to request minimal reproducible examples, error/stacktrace, clarify OS options, and include questions about LiteLLM usage and specific model details.
PiperOrigin-RevId: 806435953
The new test verifies that `output_audio_transcription` and `input_audio_transcription` attributes are unique to each `RunConfig` instance, preventing unintended side effects from modifying one instance.
PiperOrigin-RevId: 806405671
Switched the active model from `gemini-live-2.5-flash-preview` (for AI Studio) to `gemini-2.0-flash-live-preview-04-09` (for Vertex).
PiperOrigin-RevId: 806348640
Both are valid YAML, just with indent, it's more visually friend to see the data structure hierarchy.
Before
```
items:
- item1
- item2
- item3
```
After
```
items:
- item1
- item2
- item3
```
PiperOrigin-RevId: 806117290
The old live/bidi agents are using a cache to store context/history during agent transfer etc. As we have added support for session for live/bidi, we are now migrating the context/history cache to it. This improves scalability, efficiency and maintainability.
It introduces several changes:
* AudioTranscriber support is removed as now we are using native transcription from models.
* Transcription is returned as input_transcription/output_transcription fields and no longer as contents.
* We will return a new event with artifact references of file type of audio/pcm.(in addition to existing audio response event. So the users of this api need to do proper filtering here.)
PiperOrigin-RevId: 805997675
For advanced eval use cases, we do expect agent developers to have rubrics that are specific to an Eval Case and in some cases even specific to a single invocation/turn in the eval case conversation.
A separate PR will be created to consume this data model changes in ADK Eval.
PiperOrigin-RevId: 805588808
a. dump the discussion content to a tmp file first to avoid github redaction of environment variable
b. instruct the agent to use get_discussion_and_comments only when discussion content json is not available.
PiperOrigin-RevId: 805581573
Changes references from `gemini-1.5-flash` and `gemini-1.5-pro` to `gemini-2.5-flash` and `gemini-2.5-pro` in docstrings, default values, sample agents, and tests.
PiperOrigin-RevId: 805536434
**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._
console.log(`PR #${prNumber} not found or inaccessible:`, error.message);
continue;
}
// Only close if PR is still open
if (pr.data.state !== 'open') {
console.log(`PR #${prNumber} is already ${pr.data.state}, skipping`);
continue;
}
const author = pr.data.user.login;
try {
// Add comment with commit reference
await github.rest.issues.createComment({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: prNumber,
body: `Thank you @${author} for your contribution! 🎉\n\nYour changes have been successfully imported and merged via Copybara in commit ${commitSha}.\n\nClosing this PR as the changes are now in the main branch.`
This document provides context for the Gemini CLI and Gemini Code Assist to understand the project and assist with development.
This document provides context for AI coding assistants (Claude Code, Gemini CLI, GitHub Copilot, Cursor, etc.) to understand the ADK Python project and assist with development.
## Project Overview
The Agent Development Kit (ADK) is an open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control. 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.
- **Runner** - Execution engine that orchestrates the "Reason-Act" loop, manages LLM calls, executes tools, and handles multi-agent coordination
- **Tool** - Functions/capabilities agents can call (Python functions, OpenAPI specs, MCP tools, Google API tools)
- **Session** - Conversation state management (in-memory, Vertex AI, Spanner-backed)
- **Memory** - Long-term recall across sessions
## Project Architecture
Please refer to [ADK Project Overview and Architecture](https://github.com/google/adk-python/blob/main/contributing/adk_project_overview_and_architecture.md) for details.
├── unittests/ # 2600+ unit tests across 236+ files
│ ├── agents/
│ ├── tools/
│ ├── models/
│ ├── evaluation/
│ ├── a2a/
│ └── ...
└── integration/ # Integration tests
```
### ADK Live (Bidi-streaming)
- ADK live feature can be accessed from runner.run_live(...) and corresponding FAST api endpoint.
@@ -17,8 +61,132 @@ Please refer to [ADK Project Overview and Architecture](https://github.com/googl
- ADK live related configs are in [run_config.py](https://github.com/google/adk-python/blob/main/src/google/adk/agents/run_config.py).
- ADK live under multi-agent scenario: we convert the audio into text. This text will be passed to next agent as context.
- Most logics are in [base_llm_flow.py](https://github.com/google/adk-python/blob/main/src/google/adk/flows/llm_flows/base_llm_flow.py) and [gemini_llm_connection.py](https://github.com/google/adk-python/blob/main/src/google/adk/models/gemini_llm_connection.py).
- Input transcription and output transcription should be added to session as Event.
- User audio or model audio should be saved into artifacts with a reference in Event to it.
- Tests are in [tests/unittests/streaming](https://github.com/google/adk-python/tree/main/tests/unittests/streaming).
### Agent Structure Convention (Required)
**All agent directories must follow this structure:**
```
my_agent/
├── __init__.py # MUST contain: from . import agent
└── agent.py # MUST define: root_agent = Agent(...) OR app = App(...)
```
**Choose one pattern based on your needs:**
**Option 1 - Simple Agent (for basic agents without plugins):**
```python
fromgoogle.adk.agentsimportAgent
fromgoogle.adk.toolsimportgoogle_search
root_agent=Agent(
name="search_assistant",
model="gemini-2.5-flash",
instruction="You are a helpful assistant.",
description="An assistant that can search the web.",
tools=[google_search]
)
```
**Option 2 - App Pattern (when you need plugins, event compaction, custom configuration):**
```python
fromgoogle.adkimportAgent
fromgoogle.adk.appsimportApp
fromgoogle.adk.pluginsimportContextFilterPlugin
root_agent=Agent(
name="my_agent",
model="gemini-2.5-flash",
instruction="You are a helpful assistant.",
tools=[...],
)
app=App(
name="my_app",
root_agent=root_agent,
plugins=[
ContextFilterPlugin(num_invocations_to_keep=3),
],
)
```
**Rationale:** This structure allows the ADK CLI (`adk web`, `adk run`, etc.) to automatically discover and load agents without additional configuration.
## Development Setup
### Requirements
**Minimum requirements:**
- Python 3.9+ (**Python 3.11+ strongly recommended** for best performance)
-`uv` package manager (**required** - faster than pip/venv)
**Install uv if not already installed:**
```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```
### Setup Instructions
**Standard setup for development:**
```bash
# Create virtual environment with Python 3.11
uv venv --python "python3.11"".venv"
source .venv/bin/activate
# Install all dependencies for development
uv sync --all-extras
```
**Minimal setup for testing only (matches CI):**
```bash
uv sync --extra test --extra eval --extra a2a
```
**Virtual Environment Usage (Required):**
- **Always use** `.venv/bin/python` or `.venv/bin/pytest` directly
- **Or activate** with `source .venv/bin/activate` before running commands
- **Never use** `python -m venv` - always create with `uv venv` if missing
**Rationale:**`uv` is significantly faster and ensures consistent dependency resolution across the team.
**Use real code over mocks:** ADK tests should use real implementations as much as possible instead of mocking. Only mock external dependencies like network calls or cloud services.
**Test interface behavior, not implementation details:** Tests should verify that the public API behaves correctly, not how it's implemented internally. This makes tests resilient to refactoring and ensures the contract with users remains intact.
**Test Requirements:**
- Fast and isolated tests where possible
- Use real ADK components; mock only external dependencies (LLM APIs, cloud services, etc.)
- Focus on testing public interfaces and behavior, not internal implementation
- Descriptive test names that explain what behavior is being tested
- High coverage for new features, edge cases, and error conditions
- Location: `tests/unittests/` following source structure
## Docstring and comments
### Comments - Explaining the Why, Not the What
@@ -211,9 +432,49 @@ The following changes are considered breaking and necessitate a MAJOR version
- Dependency Removal: Removing support for a previously integrated third-party
library or tool type.
## Commit Message Format
## Commit Message Format (Required)
- Please use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/)
format.
- If it's not a breaking change, please add #non-breaking tag. If it's a
* Support for multiple agent types (LLM, Sequential, Parallel, Loop, Workflow)
* Agent tool support with nested agent tools
* Built-in and custom tool integration
* Callback management for all ADK callback types (before/after agent, model, tool)
* Assistant to help you build your agents with natural language
* Assistant proposes and writes agent configuration yaml files for you
* Save to test with chat interfaces as normal
* Build and debug at the same time in adk web!
* **[Core]**
* Add support for extracting cache-related token counts from LiteLLM usage ([4f85e86](https://github.com/google/adk-python/commit/4f85e86fc3915f0e67312a39fe22451968d4f1b1))
* Expose the Python code run by the code interpreter in the logs ([a2c6a8a](https://github.com/google/adk-python/commit/a2c6a8a85cf4f556e9dacfe46cf384d13d964208))
* Add run_debug() helper method for quick agent experimentation ([0487eea](https://github.com/google/adk-python/commit/0487eea2abcd05d7efd123962d17b8c6c9a9d975))
* Allow injecting a custom Runner into `agent_to_a2a` ([156d235](https://github.com/google/adk-python/commit/156d23547915e8f7f5c6ba55e0362f4b133c3968))
* Support MCP prompts via the McpInstructionProvider class ([88032cf](https://github.com/google/adk-python/commit/88032cf5c56bb2d81842353605f9f5ab4b2206ff))
* **[Models]**
* Add model tracking to LiteLlm and introduce a LiteLLM with fallbacks demo ([d4c63fc](https://github.com/google/adk-python/commit/d4c63fc5629e7d70ad8b8185be09243a01e3428f))
* Add ApigeeLlm as a model that lets ADK Agent developers to connect with an Apigee proxy ([87dcb3f](https://github.com/google/adk-python/commit/87dcb3f7ba344a2ba7d9edfc4817c9e792d90bfc))
* **[Integrations]**
* Add example and fix for loading and upgrading old ADK session databases ([338c3c8](https://github.com/google/adk-python/commit/338c3c89c6bce7f3406f729013cedcd78b809a56))
* Add support for specifying logging level for adk eval cli command ([b1ff85f](https://github.com/google/adk-python/commit/b1ff85fb2347e3402eedd42e3673be7093a99548))
* Propagate LiteLLM finish_reason to LlmResponse for use in callbacks ([71aa564](https://github.com/google/adk-python/commit/71aa5645f6c3d91fd0e0ddb1ed564188c6727080))
* Allow LLM request to override the model used in the generate content async method in LiteLLM ([ce8f674](https://github.com/google/adk-python/commit/ce8f674a287368439ba11be3285902671e9bc75a))
* Add api key argument to Vertex Session and Memory services for Express Mode support ([9014a84](https://github.com/google/adk-python/commit/9014a849eab9f77b82db4a7f2053fb2a96282f03))
* Added support for enums as arguments for function tools ([240ef5b](https://github.com/google/adk-python/commit/240ef5beea9389911e8c03a6039b353befc716ac))
* Implement artifact_version related methods in GcsArtifactService ([e194ebb](https://github.com/google/adk-python/commit/e194ebb33c62bc40403ea852a88f77a9511b61a4))
* **[Services]**
* Add support for Vertex AI Express Mode when deploying to Agent Engine ([d4b2a8b](https://github.com/google/adk-python/commit/d4b2a8b49f98a9991cb44ac7ec6e538b81a08664))
* Remove custom polling logic for Vertex AI Session Service since LRO polling is supported in express mode ([546c2a6](https://github.com/google/adk-python/commit/546c2a68165f54e694664d5b6b6740566301782b))
* Make VertexAiSessionService fully asynchronous ([f7e2a7a](https://github.com/google/adk-python/commit/f7e2a7a40ef248dd6fbba9669503b0828a12f0cc))
* Extend Bigquery detect_anomalies tool to support future data anomaly detection ([38ea749](https://github.com/google/adk-python/commit/38ea749c9cec8e65f5e768f49fd2de79b5545571))
* Add get_job_info tool to BigQuery toolset ([6429457](https://github.com/google/adk-python/commit/64294572c1c93590aa3c221015a5cb9b440ee948))
* **[Evals]**
* Add "final_session_state" to the EvalCase data model ([2274c4f](https://github.com/google/adk-python/commit/2274c4f3040b20da3690aa03272155776ca330c1))
* Marked expected_invocation as optional field on evaluator interface ([b17c8f1](https://github.com/google/adk-python/commit/b17c8f19e5fc67180d1bdc621f84cd43e357571c))
* Adds LLM-backed user simulator ([54c4ecc](https://github.com/google/adk-python/commit/54c4ecc73381cffa51cff01c7fb8a2ac59308c53))
* **[Observability]**
* Add BigQueryLoggingPlugin for event logging to BigQuery ([b7dbfed](https://github.com/google/adk-python/commit/b7dbfed4a3d4a0165e2c6e51594d1f547bec89d3))
* **[Live]**
* Add token usage to live events for bidi streaming ([6e5c0eb](https://github.com/google/adk-python/commit/6e5c0eb6e0474f5b908eb9df20328e7da85ebed9))
### Bug Fixes
* Reduce logging spam for MCP tools without authentication ([11571c3](https://github.com/google/adk-python/commit/11571c37ab948d43cbaa3a1d82522256dfe4d467))
* Fix typo in several files ([d2888a3](https://github.com/google/adk-python/commit/d2888a3766b87df2baaaa1a67a2235b1b80f138f))
* Disable SetModelResponseTool workaround for Vertex AI Gemini 2+ models ([6a94af2](https://github.com/google/adk-python/commit/6a94af24bf3367c05a5d405b7e7b79810a1fac4e))
* Bug when callback_context_invocation_context is missing in GlobalInstructionPlugin ([f81ebdb](https://github.com/google/adk-python/commit/f81ebdb622211031945eb06c3f00ff5208d94f9b))
* Support models slash prefix in model name extraction ([8dff850](https://github.com/google/adk-python/commit/8dff85099d67623dd6f4a707fb932ea55b8aaf9b))
* Do not consider events with state delta and no content as final response ([1ee93c8](https://github.com/google/adk-python/commit/1ee93c8bcb7ccd6f33658dc76b2095dd7e58aac9))
* Parameter filtering for CrewAI functions with **kwargs ([74a3500](https://github.com/google/adk-python/commit/74a3500fc5d4b07e80f914d83a0d91face28086c))
* Do not treat FinishReason.STOP as error case for LLM responses containing candidates with empty contents ([2f72ceb](https://github.com/google/adk-python/commit/2f72ceb49b452c5a1f257bce6adb004fa5d54472))
* Fixes null check for reflect_retry plugin sample ([86f0155](https://github.com/google/adk-python/commit/86f01550bd1b52d6d160e8bc54cecc6c4fe8611c))
* Creates evalset directory on evalset create ([6c3882f](https://github.com/google/adk-python/commit/6c3882f2d66f169d393171be280b6e6218b52a7c))
* Add ADK_DISABLE_LOAD_DOTENV environment variable that disables automatic loading of .env when running ADK cli, if set to true or 1 ([15afbcd](https://github.com/google/adk-python/commit/15afbcd1587d4102a4dc5c07c0c493917df9d6ea))
* Output file uploading to artifact service should handle both base64 encoded and raw bytes ([496f8cd](https://github.com/google/adk-python/commit/496f8cd6bb36d3ba333d7ab1e94e7796d2960300))
* Correct message part ordering in A2A history ([5eca72f](https://github.com/google/adk-python/commit/5eca72f9bfd05c7c28a3d738391138a59a31167d))
* Change instruction insertion to respect tool call/response pairs ([1e6a9da](https://github.com/google/adk-python/commit/1e6a9daa63050936ab421f1f684935927aebc63e))
* DynamicPickleType to support MySQL dialect ([fc15c9a](https://github.com/google/adk-python/commit/fc15c9a0c3c043c0a61dce625b8cd1ee121b4baf))
* Enable usage metadata in LiteLLM streaming ([f9569bb](https://github.com/google/adk-python/commit/f9569bbb1afbc7f0e8b6e68599590471fd112b9f))
* Fix issue with MCP tools throwing an error ([1a4261a](https://github.com/google/adk-python/commit/1a4261ad4b66cdeb39d39110a086bd6112b17516))
* Remove redundant `format` field from LiteLLM content objects ([489c39d](https://github.com/google/adk-python/commit/489c39db01465e38ecbc2c7f32781c349b8cddc9))
* Update the contribution analysis tool to use original write mode ([54db3d4](https://github.com/google/adk-python/commit/54db3d4434e0706b83a589fa2499d11d439a6e4e))
### Improvements
* Add Community Repo section to README ([432d30a](https://github.com/google/adk-python/commit/432d30af486329aa83f89c5d5752749a85c0b843))
* Undo adding MCP tools output schema to FunctionDeclaration ([92a7d19](https://github.com/google/adk-python/commit/92a7d1957367d498de773761edd142d8c108d751))
* Refactor ADK README for clarity and consistency ([b0017ae](https://github.com/google/adk-python/commit/b0017aed4472c73c3b07e71f1d65ae97a5293547))
* Add support for reversed proxy in adk web ([a0df75b](https://github.com/google/adk-python/commit/a0df75b6fa35d837086decb8802dbf1c0a6637ad))
* Avoid rendering empty columns as part of detailed results rendering of eval results ([5cb35db](https://github.com/google/adk-python/commit/5cb35db921bf86b5ad0012046bd19fa7cc1e6abb))
* Clear the behavior of disallow_transfer_to_parent ([48ddd07](https://github.com/google/adk-python/commit/48ddd078941f9240b10f052b6de171c310bc2bc6))
* Disable the scheduled execution for issue triage workflow ([a02f321](https://github.com/google/adk-python/commit/a02f321f1bdb8be9ad1873db804e0e8393268dc3))
* Include delimiter when matching events from parent nodes in content processor ([b8a2b6c](https://github.com/google/adk-python/commit/b8a2b6c57080ae29d7a02df7d9fcc2f961d422d2))
* Implement ADK-based agent factory for Tau-bench ([c0c67c8](https://github.com/google/adk-python/commit/c0c67c8698d70ddb9ed958416661f232ef9a5ed8))
* Add util to run ADK LLM Agent with simulation environment ([87f415a](https://github.com/google/adk-python/commit/87f415a7c36a1f3b6ab84d1fe939726c6ef7f34e))
* Demonstrate CodeExecutor customization for environment setup ([8eeff35](https://github.com/google/adk-python/commit/8eeff35b35d7e1538a5c9662cc8369f6ff7962f8))
* Add sample agent for VertexAiCodeExecutor ([edfe553](https://github.com/google/adk-python/commit/edfe5539421d196ca4da14d3a37fac7b598f8c8d))
* Adds a new sample agent that demonstrates how to integrate PostgreSQL databases using the Model Context Protocol (MCP) ([45a2168](https://github.com/google/adk-python/commit/45a2168e0e6773e595ecfb825d7e4ab0a38c3a38))
* Add example for using ADK with Fast MCP sampling ([d3796f9](https://github.com/google/adk-python/commit/d3796f9b33251d28d05e6701f11e80f02a2a49e1))
* 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))
* Add the ability to rewind a session to before a previous invocation ([9dce06f](https://github.com/google/adk-python/commit/9dce06f9b00259ec42241df4f6638955e783a9d1))
* Support resuming a parallel agent with multiple branches paused on tool confirmation requests ([9939e0b](https://github.com/google/adk-python/commit/9939e0b087094038b90d86c2fd35c26dd63f1157))
* Support content union as static instruction ([cc24d61](https://github.com/google/adk-python/commit/cc24d616f80c0eba2b09239b621cf3d176f144ea))
* **[Evals]**
* ADK cli allows developers to create an eval set and add an eval case ([ae139bb](https://github.com/google/adk-python/commit/ae139bb461c2e7c6be154b04f3f2c80919808d31))
* **[Integrations]**
* Allow custom request and event converters in A2aAgentExecutor ([a17f3b2](https://github.com/google/adk-python/commit/a17f3b2e6d2d48c433b42e27763f3d6df80243ca))
* **[Observability]**
* Env variable for disabling llm_request and llm_response in spans ([e50f05a](https://github.com/google/adk-python/commit/e50f05a9fc94834796876f7f112f344f788f202e))
* **[Services]**
* Allow passing extra kwargs to create_session of VertexAiSessionService ([6a5eac0](https://github.com/google/adk-python/commit/6a5eac0bdc9adc6907a28f65a3d4d7234e863049))
* Implement new methods in in-memory artifact service to support custom metadata, artifact versions, etc. ([5a543c0](https://github.com/google/adk-python/commit/5a543c00df2f7a66018df8a67efcf4ce44d4e0e4))
* Add create_time and mime_type to ArtifactVersion ([2c7a342](https://github.com/google/adk-python/commit/2c7a34259395b1294319118d0f3d1b3b867b44d6))
* Support returning all sessions when user id is none ([141318f](https://github.com/google/adk-python/commit/141318f77554ae4eb5a360bea524e98eff4a086c))
* **[Tools]**
* Support additional headers for Google API toolset ([ed37e34](https://github.com/google/adk-python/commit/ed37e343f0c997d3ee5dc98888c5e0dbd7f2a2b6))
* 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))
* Support dynamic per-request headers in MCPToolset ([6dcbb5a](https://github.com/google/adk-python/commit/6dcbb5aca642290112a7c81162b455526c15cd14))
* Add `bypass_multi_tools_limit` option to GoogleSearchTool and VertexAiSearchTool ([9a6b850](https://github.com/google/adk-python/commit/9a6b8507f06d8367488aac653efecf665619516c), [6da7274](https://github.com/google/adk-python/commit/6da727485898137948d72906d86d78b6db6331ac))
* Extend `ReflectAndRetryToolPlugin` to support hallucinating function calls ([f51380f](https://github.com/google/adk-python/commit/f51380f9ea4534591eda76bef27407c0aa7c3fae))
* Add require_confirmation param for MCP tool/toolset ([78e74b5](https://github.com/google/adk-python/commit/78e74b5bf2d895d72025a44dbcf589f543514a50))
* **[UI]**
* Granular per agent speech configuration ([409df13](https://github.com/google/adk-python/commit/409df1378f36b436139aa909fc90a9e9a0776b3a))
### Bug Fixes
* Returns dict as result from McpTool to comply with BaseTool expectations ([4df9263](https://github.com/google/adk-python/commit/4df926388b6e9ebcf517fbacf2f5532fd73b0f71))
* Fixes the identity prompt to be one line ([7d5c6b9](https://github.com/google/adk-python/commit/7d5c6b9acf0721dd230f08df919c7409eed2b7d0))
* Fix the broken langchain importing caused their 1.0.0 release ([c850da3](https://github.com/google/adk-python/commit/c850da3a07ec1441037ced1b654d8aacacd277ab))
* Fix BuiltInCodeExecutor to support visualizations ([ce3418a](https://github.com/google/adk-python/commit/ce3418a69de56570847d45f56ffe7139ab0a47aa))
* Improve error message when adk web is run in wrong directory ([4a842c5](https://github.com/google/adk-python/commit/4a842c5a1334c3ee01406f796651299589fe12ab))
* Handle App objects in eval and graph endpoints ([0b73a69](https://github.com/google/adk-python/commit/0b73a6937bd84a41f79a9ada3fc782dca1d6fb11))
* Exclude `additionalProperties` from Gemini schemas ([307896a](https://github.com/google/adk-python/commit/307896aeceeb97efed352bc0217bae10423e5da6))
* Overall eval status should be NOT_EVALUATED if no invocations were evaluated ([9fbed0b](https://github.com/google/adk-python/commit/9fbed0b15afb94ec8c0c7ab60221bbc97e481b06))
* Create context cache only when prefix matches with previous request ([9e0b1fb](https://github.com/google/adk-python/commit/9e0b1fb62b06de7ecb79bf77d54a999167d001e1))
* Handle `App` instances returned by `agent_loader.load_agent` ([847df16](https://github.com/google/adk-python/commit/847df1638cbf1686aa43e8e094121d4e23e40245))
* Add support for file URIs in LiteLLM content conversion ([85ed500](https://github.com/google/adk-python/commit/85ed500871ff55c74d16e809ddae0d4db66cbc3a))
* Only exclude scores that are None ([998264a](https://github.com/google/adk-python/commit/998264a5b1b98ac660fcc1359fb2d25c84fa0d87))
* Better handling the A2A streaming tasks ([bddc70b](https://github.com/google/adk-python/commit/bddc70b5d004ba5304fe05bcbf6e08210f0e6131))
* Correctly populate context_id in remote_a2a_agent library ([2158b3c](https://github.com/google/adk-python/commit/2158b3c91531e9125761f211f125d9ab41a55e10))
* Fix pickle data was truncated error in database session using MySql ([36c96ec](https://github.com/google/adk-python/commit/36c96ec5b356109b7c874c85d8bb24f0bf6c050d))
### Improvements
* Improve hint message in agent loader ([fe1fc75](https://github.com/google/adk-python/commit/fe1fc75c15a7983829bbe0b023f4b612b1e5c018))
* Fixes MCPToolset --> McpToolset in various places ([d4dc645](https://github.com/google/adk-python/commit/d4dc6454783f747120d407d0dc2cb78f53598d83))
* Add span for context caching handling and new cache creation ([a2d9f13](https://github.com/google/adk-python/commit/a2d9f13fa1d31e00ba9493fba321ca151cdd9366))
* Checks gemini version for `2 and above` for gemini-builtin tools ([0df6759](https://github.com/google/adk-python/commit/0df67599c0eb54a9a5df51af06483b40058953bf))
* Refactor and fix state management in the session service ([8b3ed05](https://github.com/google/adk-python/commit/8b3ed059c24903e8aca0a09d9d503b48af7df850))
* Update agent builder instructions and remove run command details ([89344da](https://github.com/google/adk-python/commit/89344da81364d921f778c8bbea93e1df6ad1097e))
* Clarify how to use adk built-in tool in instruction ([d22b8bf](https://github.com/google/adk-python/commit/d22b8bf8907e723f618dfd18e90dd0a5dbc9518c))
* Delegate the agent state reset logic to LoopAgent ([bb1ea74](https://github.com/google/adk-python/commit/bb1ea74924127d65d763a45b869da3d4ff4d5c5a))
* Adjust the instruction about default model ([214986e](https://github.com/google/adk-python/commit/214986ebeb53b2ef34c8aa37cd6403106de82c1b))
* Migrate invocation_context to callback_context ([e2072af](https://github.com/google/adk-python/commit/e2072af69f40474431b6749b7b9dc22fbcbc7730))
* Correct the callback signatures ([fa84bcb](https://github.com/google/adk-python/commit/fa84bcb5756773eadff486b99c9bd416b4faa9c6))
* Set default for `bypass_multi_tools_limit` to False for GoogleSearchTool and VertexAiSearchTool ([6da7274](https://github.com/google/adk-python/commit/6da727485898137948d72906d86d78b6db6331ac))
* Add more clear instruction to the doc updater agent about one PR for each recommended change ([b21d0a5](https://github.com/google/adk-python/commit/b21d0a50d610407be2f10b73a91274840ffdfe18))
* Add a guideline to avoid content deletion ([16b030b](https://github.com/google/adk-python/commit/16b030b2b25a9b0b489e47b4b148fc4d39aeffcb))
* Add an sample agent for the `ReflectAndRetryToolPlugin` ([9b8a4aa](https://github.com/google/adk-python/commit/9b8a4aad6fe65ef37885e5c3368d2799a2666534))
* Improve error message when adk web is run in wrong directory ([4a842c5](https://github.com/google/adk-python/commit/4a842c5a1334c3ee01406f796651299589fe12ab))
* Add an sample agent for the `ReflectAndRetryToolPlugin` ([9b8a4aa](https://github.com/google/adk-python/commit/9b8a4aad6fe65ef37885e5c3368d2799a2666534))
* Add span for context caching handling and new cache creation ([a2d9f13](https://github.com/google/adk-python/commit/a2d9f13fa1d31e00ba9493fba321ca151cdd9366))
* Disable the scheduled execution for issue triage workflow ([bae2102](https://github.com/google/adk-python/commit/bae21027d9bd7f811bed638ecce692262cb33fe5))
* Correct the callback signatures ([fa84bcb](https://github.com/google/adk-python/commit/fa84bcb5756773eadff486b99c9bd416b4faa9c6))
### Documentation
* Format README.md for samples ([0bdba30](https://github.com/google/adk-python/commit/0bdba3026345872fb907aedd1ed75e4135e58a30))
* Bump models in llms and llms-full to Gemini 2.5 ([ce46386](https://github.com/google/adk-python/commit/ce4638651f376fb6579993d8468ae57198134729))
* Announce the first ADK Community Call in the README ([731bb90](https://github.com/google/adk-python/commit/731bb9078d01359ae770719a8f5c003680ed9f3e))
* Support Oauth2 client credentials grant type ([5c6cdcd](https://github.com/google/adk-python/commit/5c6cdcd197a6780fc86d9183fa208f78c8a975d9))
* Add `ReflectRetryToolPlugin` to reflect from errors and retry with different arguments when tool errors ([e55b894](https://github.com/google/adk-python/commit/e55b8946d6a2e01aaf018d6a79d11d13c5286152))
* Support using `VertexAiSearchTool` built-in tool with other tools in the same agent ([4485379](https://github.com/google/adk-python/commit/4485379a049a5c84583a43c85d444ea1f1ba6f12))
* Support using google search built-in tool with other tools in the same agent ([d3148da](https://github.com/google/adk-python/commit/d3148dacc97f0a9a39b6d7a9640f7b7b0d6f9a6c))
* Add Rubric based tool use metric ([c984b9e](https://github.com/google/adk-python/commit/c984b9e5529b48fff64865a8b805e7e93942ea53))
* **[UI]**
* Adds `adk web` options for custom logo ([822efe0](https://github.com/google/adk-python/commit/822efe00659607bad2d19ec9a2d14c649fca2d8d))
* **[Observability]**
* **otel:** Switch CloudTraceSpanExporter to telemetry.googleapis.com ([bd76b46](https://github.com/google/adk-python/commit/bd76b46ce296409d929ae69c5c43347c73e7b365))
### Bug Fixes
* Adapt to new computer use tool name in genai sdk 1.41.0 ([c6dd444](https://github.com/google/adk-python/commit/c6dd444fc947571d089b784fde3a81e17b10cf28))
* Add AuthConfig json serialization in vertex ai session service ([636def3](https://github.com/google/adk-python/commit/636def3687a85e274e3ab44d906f6d92d49e84c0))
* Added more agent instructions for doc content changes ([7459962](https://github.com/google/adk-python/commit/745996212db156878554386be34f58658482e687))
* Convert argument to pydantic model when tool declares it accepts pydantic model as argument ([571c802](https://github.com/google/adk-python/commit/571c802fbaa80b3e65f9ce2db772b9db5a13dbc4))
* Do not re-create `App` object when loader returns an `App` ([d5c46e4](https://github.com/google/adk-python/commit/d5c46e496009eb55d78637f47162df7fcaf3a7ac))
* Fix the instruction in workflow_triage example agent ([8f3ca03](https://github.com/google/adk-python/commit/8f3ca0359e5b1306c1395770759a74aa48a52347))
* Fixes a bug that causes intermittent `pydantic` validation errors when uploading files ([e680063](https://github.com/google/adk-python/commit/e68006386fdd0da98feb9c3dce9322e44a9c914d))
* Handle A2A Task Status Update Event when streaming in remote_a2a_agent ([a5cf80b](https://github.com/google/adk-python/commit/a5cf80b952887c07bb1d56b7bdec28808edcc4a9))
* Make compactor optional in Events Compaction Config and add a default ([3f4bd67](https://github.com/google/adk-python/commit/3f4bd67b49cd60e6a2e43ccd5192efe450a6e009))
* Rename SlidingWindowCompactor to LlmEventSummarizer and refine its docstring ([f1abdb1](https://github.com/google/adk-python/commit/f1abdb1938e474564a3a76279a1a0a511f74a750))
* Rollback compaction handling from _get_contents ([84f2f41](https://github.com/google/adk-python/commit/84f2f417f77ead3748c5bbeac7f144164b9a9416))
* Set `max_output_tokens` for the agent builder ([2e2d61b](https://github.com/google/adk-python/commit/2e2d61b6fecb90cd474d6f51255678ff74b67a9b))
* Set default response modality to AUDIO in run_session ([68402bd](https://github.com/google/adk-python/commit/68402bda49083f2d56f8e8488fe13aa58b3bc18c))
* Update remote_a2a_agent to better handle streaming events and avoid duplicate responses ([8e5f361](https://github.com/google/adk-python/commit/8e5f36126498f751171bb2639c7f5a9e7dca2558))
* Update the load_artifacts tool so that the model can reliably call it for follow up questions about the same artifact ([238472d](https://github.com/google/adk-python/commit/238472d083b5aa67551bde733fc47826ff062679))
* Handle `App` instances returned by `agent_loader.load_agent` ([847df16](https://github.com/google/adk-python/commit/847df1638cbf1686aa43e8e094121d4e23e40245))
### Improvements
* Migrate VertexAiSessionService to use Agent Engine SDK ([90d4c19](https://github.com/google/adk-python/commit/90d4c19c5115c7af361effa8e12c248225ccf6ab))
* Migrate VertexAiMemoryBankService to use Agent Engine SDK ([d1efc84](https://github.com/google/adk-python/commit/d1efc8461e82fc31df940b701f1d1b5422214296), [97b950b](https://github.com/google/adk-python/commit/97b950b36b9c16467f0f42216b2dc8395346d7fe), [83fd045](https://github.com/google/adk-python/commit/83fd0457188decdabeae58b4e8be25daa89f2943))
* Add support for resolving $ref and $defs in OpenAPI schemas ([a239716](https://github.com/google/adk-python/commit/a239716930c72a0dbd2ccabeea69be46110ca48d))
* Adding the ContextFilterPlugin ([a06bf27](https://github.com/google/adk-python/commit/a06bf278cbc89f521c187ed51b032d82ffdafe2d))
* Adds plugin to save artifacts for issue [#2176](https://github.com/google/adk-python/issues/2176) ([657369c](https://github.com/google/adk-python/commit/657369cffe142ef3745cd5950d0d24a49f42f7fd))
* Expose log probs of candidates in LlmResponse ([f7bd3c1](https://github.com/google/adk-python/commit/f7bd3c111c211e880d7c1954dd4508b952704c68))
* **[Context Caching]**
* Support context caching ([c66245a](https://github.com/google/adk-python/commit/c66245a3b80192c16cb67ee3194f82c9a7c901e5))
- Support explicit context caching auto creation and lifecycle management.
* Add --otel_to_cloud experimental support ([1ae0b82](https://github.com/google/adk-python/commit/1ae0b82f5602a57ad1ca975ca0b7c85003d1a28a), [b131268](https://github.com/google/adk-python/commit/b1312680f4ea9f21c3246a1d24392619643d71f5), [7870480](https://github.com/google/adk-python/commit/7870480c63bb4fc08cfb3cabc0e1f0458f0e85bd))
* Add GenAI Instrumentation if --otel_to_cloud is enabled ([cee365a](https://github.com/google/adk-python/commit/cee365a13d0d1b1f2be046c1cc29e24a8d1fdbcc))
* Support standard OTel env variables for exporter endpoints ([f157b2e](https://github.com/google/adk-python/commit/f157b2ee4caf4055e78f4657254e45913895f5de))
* Temporarily disable Cloud Monitoring integration in --otel_to_cloud ([3b80337](https://github.com/google/adk-python/commit/3b80337faf427460e4743e25dbb92578f823513f))
* **[Services]**
* Add endpoint to generate memory from session ([2595824](https://github.com/google/adk-python/commit/25958242db890b4d2aac8612f7f7cfbb561727fa))
* **[Tools]**
* Add Google Maps Grounding Tool to ADK ([6b49391](https://github.com/google/adk-python/commit/6b493915469ecb42068e24818ab547b0856e4709))
* **MCP:** Initialize tool_name_prefix in MCPToolse ([86dea5b](https://github.com/google/adk-python/commit/86dea5b53ac305367283b7e353b60d0f4515be3b))
* **[Evals]**
* Data model for storing App Details and data model for steps ([01923a9](https://github.com/google/adk-python/commit/01923a9227895906ca8ae32712d65b178e2cd7d5))
* Adds Rubric based final response evaluator ([5a485b0](https://github.com/google/adk-python/commit/5a485b01cd64cb49735e13ebd5e7fa3da02cd85f))
* Populate AppDetails to each Invocation ([d486795](https://github.com/google/adk-python/commit/d48679582de91050ca9c5106402319be9a8ae7e8))
* **[Samples]**
* Make the bigquery sample agent run with ADC out-of-the-box ([10cf377](https://github.com/google/adk-python/commit/10cf37749417856e394e62896231e41b13420f18))
### Bug Fixes
* Close runners after running eval ([86ee6e3](https://github.com/google/adk-python/commit/86ee6e3fa3690148d60358fc3dacb0e0ab40942b))
* Filter out thought parts when saving agent output to state ([632bf8b](https://github.com/google/adk-python/commit/632bf8b0bcf18ff4e4505e4e5f4c626510f366a2))
* Ignore empty function chunk in LiteLlm streaming response ([8a92fd1](https://github.com/google/adk-python/commit/8a92fd18b600da596c22fd80c6148511a136dfd0))
* Introduces a `raw_mcp_tool` method in `McpTool` to provide direct access to the underlying MCP tool ([6158075](https://github.com/google/adk-python/commit/6158075a657f8fe0835679e509face6191905403))
* Make a copy of the `columns` instead of modifying it in place ([aef1ee9](https://github.com/google/adk-python/commit/aef1ee97a55a310f3959d475b8d7d6bc3915ae48))
* Prevent escaping of Latin characters in LLM response ([c9ea80a](https://github.com/google/adk-python/commit/c9ea80af28e586c9cc1f643b365cdba82f80c700))
* Retain the consumers and transport registry when recreating the ClientFactory in remote_a2a_agent.py ([6bd33e1](https://github.com/google/adk-python/commit/6bd33e1be36f741a6ed0514197550f9f336262ed))
* Remove unsupported 'type': 'unknown' in test_common.py for fastapi 0.117.1 ([3745221](https://github.com/google/adk-python/commit/374522197fa6843f786bfd12d17ce0fc20461dfd))
### Documentation
* Correct the documentation of `after_agent_callback` ([b9735b2](https://github.com/google/adk-python/commit/b9735b2193267645781b268231d63c23c6fec654))
* Allow users to pass their own agent card to to_a2a method [a1679da](https://github.com/google/adk-python/commit/a1679dae3fef70f1231afba3e97d45b59c314ae3)
* Allow custom part converters in A2A classes [b05fef9](https://github.com/google/adk-python/commit/b05fef9ba71f95ab2658eb4eb5608c141d49f82f)
* [Tools]
***[Tools]**
* Allow setting agent/application name and compute project for BigQuery tools [11a2ffe](https://github.com/google/adk-python/commit/11a2ffe35adbae977b49ceccf0e76e20c6dc90b6)
* Add GkeCodeExecutor for sandboxed code execution on GKE [72ff9c6](https://github.com/google/adk-python/commit/72ff9c64a291aebb50b07446378f375e58882c4e)
* Add a tool confirmation flow that can guard tool execution with explicit confirmation and custom input [a17bcbb](https://github.com/google/adk-python/commit/a17bcbb2aa0f5c6aca460db96ed1cb7dd86fef84)
* Add audience and prompt as configurable for OAuth flows [edda922](https://github.com/google/adk-python/commit/edda922791f15ac37830ed95ebf76b9f836d9db4)
* Allow user specify embedding model for file retrieval [67f23df](https://github.com/google/adk-python/commit/67f23df25ad47aff3cb36d0fc9ce2c9b97bde09b)
* [Core]
***[Core]**
* Allow all possible values for `agent_class` field in all Agent Configs [3bc2d77](https://github.com/google/adk-python/commit/3bc2d77b4d180e9c42b30d4d1ce580aa75abe501)
* Allow agent loader to load built-in agents from special directories in adk folder [578fad7](https://github.com/google/adk-python/commit/578fad7034a7b369a490ad0afa4dd2820463c22d)
* Upgrade ADK runner to use App in addition to root_agent [4df79dd](https://github.com/google/adk-python/commit/4df79dd5c92d96096d031b26470458d0bca79a79)
* Allow inject artifact into instructions [bb4cfde](https://github.com/google/adk-python/commit/bb4cfdec12370955d4038d6d8c86e04691f2308e)
* [Misc] Create an initial ADK release analyzer agent to find the doc updates needed between releases [e3422c6](https://github.com/google/adk-python/commit/e3422c616d18ec3850454ee83f2ef286198543ec)
***[Misc]** Create an initial ADK release analyzer agent to find the doc updates needed between releases [e3422c6](https://github.com/google/adk-python/commit/e3422c616d18ec3850454ee83f2ef286198543ec)
### Bug Fixes
@@ -226,7 +538,7 @@ with Bigtable for building AI Agent applications(experimental feature) ([a953807
### Improvements
* Add Github workflow config for the ADK Answering agent ([8dc0c94](https://github.com/google/adk-python/commit/8dc0c949afb9024738ff7ac1b2c19282175c3200))
* Add GitHub workflow config for the ADK Answering agent ([8dc0c94](https://github.com/google/adk-python/commit/8dc0c949afb9024738ff7ac1b2c19282175c3200))
* Import AGENT_CARD_WELL_KNOWN_PATH from adk instead of from a2a directly ([37dae9b](https://github.com/google/adk-python/commit/37dae9b631db5060770b66fce0e25cf0ffb56948))
* Make `LlmRequest.LiveConnectConfig` field default to a factory ([74589a1](https://github.com/google/adk-python/commit/74589a1db7df65e319d1ad2f0676ee0cf5d6ec1d))
* Update the prompt to make the ADK Answering Agent more objective ([2833030](https://github.com/google/adk-python/commit/283303032a174d51b8d72f14df83c794d66cb605))
@@ -285,14 +597,13 @@ with Bigtable for building AI Agent applications(experimental feature) ([a953807
* [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 a 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))
* [Services]Support passing fully qualified agent engine resource name when constructing session service and memory service ([2e77804](https://github.com/google/adk-python/commit/2e778049d0a675e458f4e35fe4104ca1298dbfcf))
* [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))
@@ -405,7 +716,7 @@ with Bigtable for building AI Agent applications(experimental feature) ([a953807
### Documentation
* Update the a2a exmaple link in README.md [d0fdfb8](https://github.com/google/adk-python/commit/d0fdfb8c8e2e32801999c81de8d8ed0be3f88e76)
* Update the a2a example link in README.md [d0fdfb8](https://github.com/google/adk-python/commit/d0fdfb8c8e2e32801999c81de8d8ed0be3f88e76)
* Adds AGENTS.md to provide relevant project context for the Gemini CLI [37108be](https://github.com/google/adk-python/commit/37108be8557e011f321de76683835448213f8515)
* Add adk project overview and architecture [28d0ea8](https://github.com/google/adk-python/commit/28d0ea876f2f8de952f1eccbc788e98e39f50cf5)
@@ -600,7 +911,7 @@ with Bigtable for building AI Agent applications(experimental feature) ([a953807
* Fix typos in README for sample bigquery_agent and oauth_calendar_agent ([9bdd813](https://github.com/google/adk-python/commit/9bdd813be15935af5c5d2a6982a2391a640cab23))
* Make tool_call one span for telemetry and renamed to execute_tool ([999a7fe](https://github.com/google/adk-python/commit/999a7fe69d511b1401b295d23ab3c2f40bccdc6f))
* Use media type in chat window. Remove isArtifactImage and isArtifactAudio reference ([1452dac](https://github.com/google/adk-python/commit/1452dacfeb6b9970284e1ddeee6c4f3cb56781f8))
* Set output_schema correctly for LiteLllm ([6157db7](https://github.com/google/adk-python/commit/6157db77f2fba4a44d075b51c83bff844027a147))
* Set output_schema correctly for LiteLlm ([6157db7](https://github.com/google/adk-python/commit/6157db77f2fba4a44d075b51c83bff844027a147))
An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
An open-source, code-first Python framework for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
</h3>
<h3 align="center">
Important Links:
@@ -22,38 +22,42 @@
</h3>
</html>
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.
Agent Development Kit (ADK) is a flexible and modular framework that applies
software development principles to AI agent creation. It is designed to
simplify building, deploying, and orchestrating agent workflows, from simple
tasks to complex systems. While optimized for Gemini, ADK is model-agnostic,
deployment-agnostic, and compatible with other frameworks.
---
## ✨ What's new
## 🔥 What's new
- **Agent Config**: Build agents without code. Check out the
- **Custom Service Registration**: Add a service registry to provide ageneric 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)).
- **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))
- **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.
- **Deploy Anywhere**: Easily containerize and deploy agents on Cloud Run or
scale seamlessly with Vertex AI Agent Engine.
## 🤖 Agent2Agent (A2A) Protocol and ADK Integration
For remote agent-to-agent communication, ADK integrates with the
Note: The development version is built directly from the latest code commits. While it includes the newest fixes and features, it may also contain experimental changes or bugs not present in the stable release. Use it primarily for testing upcoming changes or accessing critical fixes before they are officially released.
## 🤖 Agent2Agent (A2A) Protocol and ADK Integration
For remote agent-to-agent communication, ADK integrates with the
# Create parent agent and assign children via sub_agents
coordinator=LlmAgent(
name="Coordinator",
model="gemini-2.0-flash",
model="gemini-2.5-flash",
description="I coordinate greetings and tasks.",
sub_agents=[# Assign sub_agents here
greeter,
@@ -144,10 +155,20 @@ 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.
## Community Repo
We have [adk-python-community repo](https://github.com/google/adk-python-community)that is home to a growing ecosystem of community-contributed tools, third-party
service integrations, and deployment scripts that extend the core capabilities
of the ADK.
## 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.
## Community Events
- [Completed] ADK's 1st community meeting on Wednesday, October 15, 2025. Remember to [join our group](https://groups.google.com/g/adk-community) to get access to the [recording](https://drive.google.com/file/d/1rpXDq5NSH8-MyMeYI6_5pZ3Lhn0X9BQf/view), and [deck](https://docs.google.com/presentation/d/1_b8LG4xaiadbUUDzyNiapSFyxanc9ZgFdw7JQ6zmZ9Q/edit?slide=id.g384e60cdaca_0_658&resourcekey=0-tjFFv0VBQhpXBPCkZr0NOg#slide=id.g384e60cdaca_0_658).
## đź“„ License
This project is licensed under the Apache 2.0 License - see the [LICENSE](LICENSE) file for details.
@@ -99,7 +99,7 @@ Agent: âś… Great news! Your reimbursement has been approved by the manager. Proc
The human-in-the-loop process follows this pattern:
1. **Initial Call**: Root agent delegates approval request to remote approval agent for amounts >$100
2. **Pending Response**: Remote approval agent returns immediate response with `status: "pending"` and ticket ID and serface the approval request to root agent
2. **Pending Response**: Remote approval agent returns immediate response with `status: "pending"` and ticket ID and surface the approval request to root agent
3. **Agent Acknowledgment**: Root agent informs user about pending approval status
4. **Human Interaction**: Human manager interacts with root agent to review and approve/reject the request
5. **Updated Response**: Root agent receives updated tool response with approval decision and send it to remote agent
You are an intelligent Agent Builder Assistant specialized in creating and configuring ADK (Agent Development Kit) multi-agent systems using YAML configuration files.
## Your Purpose
Help users design, build, and configure sophisticated multi-agent systems for the ADK framework. You guide users through the agent creation process by asking clarifying questions, suggesting optimal architectures, and generating properly formatted YAML configuration files that comply with the ADK AgentConfig schema.
6. **ADK AgentConfig Schema Querying**: Use the query_schema to dynamically query ADK AgentConfig schema for accurate field definitions
7. **ADK Knowledge & Q&A**: Answer questions about ADK concepts, APIs, usage patterns, troubleshooting, and best practices using comprehensive research capabilities
## ADK AgentConfig Schema Information
Instead of embedding the full ADK AgentConfig schema, you have access to the `query_schema` that allows you to:
- Query ADK AgentConfig schema overview: Use query_type="overview" to get high-level structure
- Explore ADK AgentConfig schema components: Use query_type="component" with component name (e.g., "tools", "model")
- Get ADK AgentConfig schema field details: Use query_type="field" with field_path (e.g., "tools.function_tool.function_path")
- List all ADK AgentConfig schema properties: Use query_type="properties" to get comprehensive property list
Always use the query_schema tool when you need specific ADK AgentConfig schema information to ensure accuracy.
* **IF ESTABLISHED**: Use the existing session root directory - DO NOT ask again
* **IF NOT ESTABLISHED**: Ask user for root directory to establish working context
- **MODEL PREFERENCE**: Only ask for model preference when you determine that LlmAgent(s) will be needed
* **When to ask**: After analyzing requirements and deciding that LlmAgent is needed for the solution
* **DEFAULT**: Use "{default_model}" (your current model) if user doesn't specify
* **EXAMPLES**: "gemini-2.5-flash", "gemini-2.5-pro", etc.
* **RATIONALE**: Only LlmAgent requires model specification; workflow agents do not
- **CRITICAL PATH RESOLUTION**: If user provides a relative path (e.g., `./config_agents/roll_and_check`):
* **FIRST**: Call `resolve_root_directory` to get the correct absolute path
* **VERIFY**: The resolved path matches user's intended location
* **EXAMPLE**: `./config_agents/roll_and_check` should resolve to `/Users/user/Projects/adk-python/config_agents/roll_and_check`, NOT `/config_agents/roll_and_check`
- Understand the user's goals and requirements through targeted questions
- Explore existing project structure using the RESOLVED ABSOLUTE PATH
* **CRITICAL**: Use only the final component of the root folder path as project_folder_name (e.g., for `./config_based/roll_and_check`, use `roll_and_check` not `config_based.roll_and_check`)
- No function declarations in YAML (handled automatically by ADK)
**TOOL IMPLEMENTATION STRATEGY:**
- **For simple/obvious tools**: Implement them directly with actual working code
* Example: dice rolling, prime checking, basic math, file operations
* Don't ask users to "fill in TODO comments" for obvious implementations
- **For complex/business-specific tools**: Generate proper function signatures with TODO comments
* Example: API integrations requiring API keys, complex business logic
- **Always generate correct function signatures**: If user wants `roll_dice` and `is_prime`, generate those exact functions, not generic `tool_name`
**CRITICAL: Tool Usage Patterns - MANDATORY FILE TYPE SEPARATION**
⚠️ **YAML FILES (.yaml, .yml) - MUST USE CONFIG TOOLS:**
- **ALWAYS use `write_config_files`** for writing YAML configuration files (root_agent.yaml, etc.)
- **ALWAYS use `read_config_files`** for reading YAML configuration files
- **NEVER use `write_files` for YAML files** - it lacks validation and schema compliance
⚠️ **PYTHON/OTHER FILES (.py, .txt, .md) - USE GENERAL FILE TOOLS:**
- **Use `write_files`** for Python tools, scripts, documentation, etc.
- **Use `read_files`** for non-YAML content
⚠️ **WHY THIS SEPARATION MATTERS:**
- `write_config_files` validates YAML syntax and ADK AgentConfig schema compliance
- `write_files` is raw file writing without validation
- Using wrong tool can create invalid configurations
- **For ADK code questions**: Use `search_adk_source` then `read_files` for complete context
- **File deletion**: Use `delete_files` for multiple file deletion with backup options
**TOOL GENERATION RULES:**
- **Match user requirements exactly**: Generate the specific functions requested
- **Use proper parameter types**: Don't use generic `parameter: str` when specific types are needed
- **Implement when possible**: Write actual working code for simple, well-defined functions
- **ONE TOOL PER FILE POLICY**: Always create separate files for individual tools
* **Example**: Create `roll_dice.py` and `is_prime.py` instead of `dice_tools.py`
* **Benefit**: Enables easy cleanup when tools are no longer needed
* **Exception**: Only use multi-tool files for legitimate toolsets with shared logic
### 4. Validation Phase
- Review generated configurations for schema compliance
- Test basic functionality when possible
- Provide clear next steps for the user
## Available Tools
You have access to comprehensive tools for:
- **Configuration Management**: Read/write multiple YAML configs with validation and schema compliance
- **File Management**: Read/write multiple files (Python tools, scripts, documentation) with full content handling
- **Project Exploration**: Analyze directory structures and suggest file locations
- **Schema Exploration**: Query AgentConfig schema dynamically for accurate field information
- **ADK Source Search**: Search ADK source code with regex patterns for precise code lookups
- **ADK Knowledge**: Research ADK concepts using local source search and web-based tools
- **Research**: Search GitHub examples and fetch relevant code samples
- **Working Directory**: Resolve paths and maintain context
### When to Use Research Tools
**ALWAYS use research tools when:**
1. **User asks ADK questions**: Any questions about ADK concepts, APIs, usage patterns, or troubleshooting
2. **Unfamiliar ADK features**: When user requests features you're not certain about
3. **Agent type clarification**: When unsure about agent types, their capabilities, or configuration
4. **Best practices**: When user asks for examples or best practices
5. **Error troubleshooting**: When helping debug ADK-related issues
6. **Agent building uncertainty**: When unsure how to create agents or what's the best practice
7. **Architecture decisions**: When evaluating different approaches or patterns for agent design
**Research Tool Usage Patterns:**
**For ADK Code Questions (NEW - Preferred Method):**
1. **search_adk_source** - Find exact code patterns with regex
2. **read_files** - Get complete file context for detailed analysis
3. **query_schema** - Query AgentConfig schema for field definitions
**For External Examples and Documentation:**
- **google_search_agent**: Search and analyze web content (returns full page content, not just URLs)
* Search within key repositories: "site:github.com/google/adk-python ADK SequentialAgent examples"
* General searches: "ADK workflow patterns", "ADK tool integration patterns"
* Returns complete page content as search results - no need for additional URL fetching
- **url_context_agent**: Fetch specific URLs only when:
* Specific URLs are mentioned in search results that need additional content
* User provides specific URLs in their query
* You need to fetch content from URLs found within google_search results
* NOT needed for general searches - google_search_agent already provides page content
**Research for Agent Building:**
- When user requests complex multi-agent systems: Search for similar patterns in samples
- When unsure about tool integration: Look for tool usage examples in contributing/samples
- When designing workflows: Find SequentialAgent, ParallelAgent, or LoopAgent examples
- When user needs specific integrations: Search for API, database, or service integration examples
## Code Generation Guidelines
### When Creating Python Tools or Callbacks:
1. **Always search for current examples first**: Use google_search_agent to find "ADK tool_context examples" or "ADK callback_context examples"
2. **Reference contributing/samples**: Use google_search_agent to find examples, or url_context_agent only if specific URLs are identified that need additional content
3. **Look for similar patterns**: Search for tools or callbacks that match your use case
4. **Use snake_case**: Function names should be snake_case (e.g., `check_prime`, `roll_dice`)
5. **Remove tool suffix**: Don't add "_tool" to function names
6. **Implement simple functions**: For obvious functions like `is_prime`, `roll_dice`, replace TODO with actual implementation
7. **Keep TODO for complex**: For complex business logic, leave TODO comments
8. **Follow current ADK patterns**: Always search for and reference the latest examples from contributing/samples
### Research and Examples:
- Use google_search_agent to find "ADK [use-case] examples" or "ADK [pattern] configuration" (returns full content)
- Use url_context_agent only when:
* Specific URLs are found in search results that need additional content
* User provides specific URLs to analyze
* You need to fetch specific examples from identified URLs:
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Jack Sun
Hangfei Lin
Yeesian Ng
Xuan Yang
Google Team Member
Nimantha Cooray
Hyeongjun Kang
Kathy Wu
Eliza Huang
Yifan Wang
George Weale
Shan Cao
Haegyun Lee
Josh Soref
Kacper Jawoszek
G. Hussain Chinoy
nikkie
Jeroen Overschie
FingerLiu
Ankur Sharma
Lavi Nigam
Dongyu Jia
Ayush Agrawal
Kevin Qian
Jaroslav Pantsjoha
Shubham Saboo
Ieva Grublyte
Ayam
hung12ct
Vishwa Murugan
Omar Elcircevi
Xinran (Sherry) Tang
Artem Oroberts
Liang Wu
SOORAJ TS
Shangjie Chen
Hoonji Baek
Marlin Ranasinghe
Max Ind
Keyur Joshi
Al Duncanson
Vrajesh Babu A V
Alexey Guseynov
Luis Pabon
Parham MohammadAlizadeh
ejfn
Xiang (Sean) Zhou
Giovanni Galloro
Doug Gebert
Lin Nikaido
machache
Joseph Pagadora
Joe Fernandez
Haoming Chen
Che Liu
Douglas Reid
Bo Yang
Sasha Sobran
shsha4
TanejaAnkisetty
Bastien Jacot-Guillarmod
Nikhil Purwant
guillaume blaquiere
Kel Markert
Afonso Menegola