feat: multi agent sept 2025 #376
Draft
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
docs: Document update
chore: Reuse Ai Foundry for MACAE
docs: added missing parameter
Updated link
docs: Update re-use-log-analytics.md for link update
…based on existing and new deployment
fix: Bug fix # 20359
Centralizes Azure configuration and access token retrieval in AppConfig, removing duplicate logic from JsonService and v3/config/settings.py. Updates environment sample and settings to use new config properties and methods for improved maintainability and consistency.
Moved Azure credential logic from common.auth.azure_credential_utils.py into AppConfig, removing the utility module and updating all references to use AppConfig's methods. Deleted the CosmosMemoryContext implementation in context/cosmos_memory_kernel.py. Cleaned up unused imports and updated credential usage in affected modules for consistency.
Refactored the format_dates_in_messages function from app_kernel.py to common/utils/utils_date.py for better modularity and reuse. Updated imports to reflect the new location.
Introduces a new 'services' module under v3/common with BaseAPIService, MCPService, and FoundryService for async HTTP and Azure AI Foundry interactions. Updates AppConfig to support an optional MCP server endpoint and fixes credential method references. These changes provide a foundation for modular service access and easier integration with MCP and Foundry APIs.
Renamed JsonService to TeamService and moved the file to src/backend/v3/common/services/team_service.py. Updated all references in router.py to use TeamService instead of JsonService for team configuration operations.
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Replaces the call to self.get_access_token() with config.get_access_token() in the TeamService class. This change ensures the access token is retrieved from the config object, possibly improving consistency or fixing a bug related to token acquisition.
Removed the '/plan/:planId/create' route and updated navigation to use '/plan/:planId' after plan creation. Made 'team_id' mandatory in InputTask, cleaned up code formatting, and clarified comments. This streamlines the plan creation process and ensures consistent routing and data requirements.
Moved the Azure model deployment listing logic from TeamService to FoundryService for better separation of concerns and code reuse. Updated TeamService to use FoundryService for model deployment validation and status summary. Improved logging and configuration handling in FoundryService.
Upgraded Python requirement to 3.10 and updated several dependencies in pyproject.toml. Refactored authentication to use JWTVerifier and adjusted related imports. Removed unused __init__.py files from multiple modules to clean up the package structure.
Introduces abstract methods for adding, updating, retrieving, and deleting team configurations in the DatabaseBase class. Also adds a convenience alias for getting steps by plan for compatibility.
Replaces the _get_or_create_container method with _get_container, removing container creation and only retrieving the container client. Updates error logging for clarity. This simplifies container management and assumes the container already exists.
Moved all mcp_server files from src/backend/v3/mcp_server/ to src/mcp_server/, updating imports and package structure accordingly. Added src/mcp_server/__init__.py to define the package and version. Updated pyproject.toml script entry point. Also removed a redundant utils/__init__.py file.
Introduces methods in TeamService to persist and retrieve the selected team using localStorage. HomePage now loads the default team from storage if available, and saves the selected team after determining it. Also adds a check to prevent creating a new task if no team is selected.
Renamed several frontend files from .ts to .tsx for consistency and future React component support. Moved RAIErrorCard.css to styles directory and updated import path. Added deployment_name field to agent definitions in default_team.json. Added a console.log for formData in TeamService for debugging.
The RAI content safety check in the backend plan creation endpoint has been commented out, allowing requests to bypass this validation. In the frontend HomePage, the team loading state is now correctly set to false when a default team is loaded from storage, improving UI responsiveness.
Eliminated onInputSubmit and onQuickTaskSelect props from HomeInput and related model. Also removed the handleNewTask function and its usage in HomePage, as these props and logic are no longer needed.
Refactored the iconMap definition from HomeInput.tsx to homeInput.tsx to centralize icon mapping logic and enable reuse across components. Updated imports and usage accordingly.
Added the team_id field to the Plan data model and updated the create_plan endpoint to require team_id, returning an error if missing. Also updated CosmosDB queries to include user_id and serialize datetime fields. Removed the /generate_plan endpoint from the router.
Simplifies team deletion logic in CosmosDBClient by removing redundant methods and partition key strategies. Adds session_id as a required field in TeamConfiguration and ensures session_id is generated and used consistently in team creation and deletion. Updates TeamService to use the new deletion method.
Introduces get_data_by_type_and_session_id to DatabaseBase and implements it in CosmosDBClient. This method allows querying documents by data_type, session_id, and user_id, improving data retrieval granularity.
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.
Purpose
Does this introduce a breaking change?
How to Test
What to Check
Verify that the following are valid
Other Information