fix: Aggregate state_delta from parallel AgentTool calls #1155
Conversation
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
Previously, when a root agent executed multiple sub-agents (AgentTools) in parallel, the `state_delta` from each sub-agent was not correctly aggregated. The `merge_parallel_function_response_events` function would often result in the session state reflecting only the output from one of the sub-agents, as their `state_delta`s (containing `output_key` results) could overwrite each other. This commit updates `merge_parallel_function_response_events` in `src/google/adk/flows/llm_flows/functions.py` to properly initialize and update a `merged_state_delta` dictionary. It iterates through each function response event from parallel tool calls and updates this dictionary with each sub-agent's `state_delta`. This ensures that all `output_key` results from parallel AgentTool calls are preserved and correctly reflected in the final session state. A new parallel_agent_tool_state_merge sample has been added to contributing/samples. Fixes google#1153
ystory
force-pushed
the
fix/1153-parallel-state-delta-aggregation-
branch
from
1d3ebb5 to
4ef2177
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Previously, when a root agent executed multiple sub-agents (AgentTools) in parallel, the
state_deltafrom each sub-agent was not correctly aggregated. Themerge_parallel_function_response_eventsfunction would often result in the session state reflecting only the output from one of the sub-agents, as theirstate_deltas (containingoutput_keyresults) could overwrite each other.Before fix:
This commit updates
merge_parallel_function_response_eventsinsrc/google/adk/flows/llm_flows/functions.pyto properly initialize and update amerged_state_deltadictionary. It iterates through each function response event from parallel tool calls and updates this dictionary with each sub-agent'sstate_delta. This ensures that alloutput_keyresults from parallel AgentTool calls are preserved and correctly reflected in the final session state.After fix:
A new parallel_agent_tool_state_merge sample has been added to contributing/samples.
Fixes #1153