Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
…ypass path to DiscoveryEngineSearchTool When bypass_multi_tools_limit=True converts VertexAiSearchTool to a client-side DiscoveryEngineSearchTool, the resulting tool was always named 'discovery_engine_search' — an internal implementation name. Developers had no way to express a domain-specific tool name, forcing them to leak GCP internals into agent instructions or accept runtime ValueError: Tool 'search' not found crashes. DiscoveryEngineSearchTool already accepted name/description params but they were unreachable through the conversion path. - Add optional name/description kwargs to VertexAiSearchTool.__init__ stored as _bypass_tool_name/_bypass_tool_description to never shadow the base-class 'vertex_ai_search' name used in the grounding path - Forward them in llm_agent.py during the bypass conversion - Add 5 unit tests for the agent conversion path - Add 4 unit tests for DiscoveryEngineSearchTool constructor Fixes: google#7100 (partial — addresses problem 3, hardcoded tool naming) Related: google#7101
musi22
force-pushed
the
fix-search-tool-custom-naming
branch
from
September 15, 2026 17:58
18b4a3c to
a25cde5
Compare
Author
|
@googlebot I signed it! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Addresses Problem 3 identified in #7100.
When
bypass_multi_tools_limit=Trueis enabled onVertexAiSearchTool, ADK converts the tool to a client-sideDiscoveryEngineSearchToolduring canonical tool resolution. Previously, this conversion always named the resulting tool'discovery_engine_search'-- an internal implementation detail.This caused two key problems for developers:
'discovery_engine_search'explicitly in prompts rather than domain concepts like'product_search'or'knowledge_base'.searchorquery) instead ofdiscovery_engine_search, leading toValueError: Tool '...' not found.While
DiscoveryEngineSearchToolalready supportednameanddescriptionparameters, they were unreachable through theVertexAiSearchToolconversion path.Changes
nameanddescriptionparameters toVertexAiSearchTool.__init__._bypass_tool_nameand_bypass_tool_descriptionto avoid shadowingBaseTool.name = 'vertex_ai_search'(preserving backward compatibility with the built-in grounding path).nameanddescriptionduring bypass conversion inllm_agent.py.tests/unittests/agents/test_llm_agent_fields.py.tests/unittests/tools/test_discovery_engine_search_tool.py.Related Issues
bypass_multi_tools_limitonVertexAiSearchToolis a leaky abstraction: unexpected dependencies, silent shift from inbuilt RAG to tool calling, and prompt fragility #7100 (Problem 3: Hardcoded tool name)google-cloud-discoveryengineimport errors)Testing Plan
DiscoveryEngineSearchTool(pytest tests/unittests/tools/test_discovery_engine_search_tool.py) -- 31/31 passed.-[x] Ran unit tests for VAIS tool conversion (
pytest tests/unittests/agents/test_llm_agent_fields.py -k "vais") -- 10/10 passed.