Testing #1062
Merged
Testing #1062
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
- Added web_search_filters parameter to support restricting web searches to specific allowed domains - Updated OpenAI response handler to configure web search tool with domain filters when provided - Modified request parsing and configuration loading to pass through web search filter settings - Extended BaseService to store web search filter parameters - Maintained backwards compatibility by using web_search_preview when no filters are specified
feat: add domain filtering for web search tool
* feat: add cost tracking and usage limits for bridges, folders and API keys * remove bridge data also with both vID and bID --------- Co-authored-by: Prayanshrajput <prayansh75@gmail.com> Co-authored-by: Husain Baghwala <89205542+Husainbw786@users.noreply.github.com>
- Removed duplicate code blocks from unresolved git merge markers - Standardized version_id parameter usage across all limit check functions - Consolidated Redis cache key generation to include version_id for proper cache isolation Co-authored-by: Husain Baghwala <husainhackerrank@gmail.com>
) - Added score threshold parameter to filter vector search results based on similarity scores - Enhanced response metadata to include per-result scores and similarity information when filtering is enabled - Modified get_vectors_and_text endpoint to support optional score-based filtering Co-authored-by: Husain Baghwala <husainhackerrank@gmail.com>
* feat: add Grok AI service integration - Integrated xAI's Grok API as a new service provider with full chat completion support - Added service configuration mappings and parameter translations for Grok - Implemented conversation formatting with support for images and tool calls * refactor: migrate Grok API integration to use custom fetch utility - Replaced httpx client with shared fetch function for consistency across API integrations - Updated token calculation to handle xAI SDK's SamplingUsage object attributes - Removed unused grok_test_model function * feat: add Grok AI service support with grok-4-fast model --------- Co-authored-by: Husain Baghwala <husainhackerrank@gmail.com>
…nd improve bridge serialization (#1039) Co-authored-by: Yogesh Patel <ygurjar932@gmail.com>
- Upgraded anthropic SDK from 0.57.1 to 0.72.0 to support web search functionality - Implemented Anthropic's official web_search_20250305 tool with optional domain filtering - Enhanced streaming response handling to properly merge text blocks and parse tool inputs including web search results Co-authored-by: Husain Baghwala <husainhackerrank@gmail.com>
- Added support for web_search_filters in bridge configuration updates - Extended validation rules to accept web_search_filters parameter - Modified update logic to process web search filter settings when provided - Preserved existing configuration handling while adding new filter capability Co-authored-by: Yogesh Patel <ygurjar932@gmail.com>
…on (#1054) - Updated the getConfiguration function to safely extract API keys for all services instead of a single service - Implemented iteration over all services to retrieve their respective API keys - Improved error logging for API key access issues
- Added doctstar_embed_token generation using org_id and DocStar configuration parameters - Included doctstar_embed_token in the API response payload for bridge configuration - Integrated with DocStar service using dedicated organization and collection IDs Co-authored-by: Yogesh Patel <ygurjar932@gmail.com>
- Changed the base URL for the AsyncOpenAI instance from 'https://backend.ai.ml/openai' to 'https://api.ai.ml/openai' to ensure correct API endpoint usage.
- Added support for image generation in the AiMl service by introducing the AiMlImageModel. - Enhanced the ai_ml_call.py to handle image requests and format responses accordingly. - Integrated image generation API calls and response handling, including error management and GCP upload. - Updated Response_formatter to accommodate image response formatting for the AiMl service.
- Implemented process_variable_state function to automatically populate variables with default values when empty or missing - Updated variable_state structure to support status, default_value, and value fields for enhanced variable management - Modified variable status determination logic to work with new nested variable_state structure
…tate_for_default_variable feat: add default value handling for variables based on variable_state
* now raises both initail as well as fallback error * with accessible error messages * string again * reqired structure * removed hardcoded value and json handling in executionhandler --------- Co-authored-by: kishan31082003 <savita.singh5477@gmail.com>
- Added logic to handle response formatting and message publishing for playground chat completions. - Introduced a channel ID construction based on organization, bridge, and version IDs. - Implemented error handling for message publishing failures. - Updated chat function to include message ID in responses and send data to playground when applicable. Co-authored-by: Husain Baghwala <husainhackerrank@gmail.com>
- Added logic to send responses for both successful and error cases in the chat function. - Improved error handling by ensuring structured error responses are sent back to the client. - Updated the response format to include relevant variables when applicable.
- Relocated sendResponse to execute after testcase_id generation - Ensures the response includes the newly generated testcase_id before sending Co-authored-by: Husain Baghwala <husainhackerrank@gmail.com>
…responses - Extracted single-bridge configuration logic into `_prepare_configuration_response` helper function - Added `_collect_connected_agent_configs` to recursively gather configurations for connected agents - Modified `getConfiguration` to return a dictionary of bridge configurations keyed by bridge ID instead of a single flat configuration object
…cation - Introduced chat_multiple_agents function to handle requests for multiple agents, improving flexibility in agent interactions. - Updated chat and image functions to utilize chat_multiple_agents, ensuring consistent handling of agent requests. - Enhanced error handling and logging for agent transfers, allowing for smoother transitions between agents. - Modified various service functions to support bridge configurations, enabling better management of agent-specific settings.
feat: add user and bridge metadata to error webhooks and update folde…
- Moved transfer action type check to caller for better control flow - Removed redundant bridge configuration validation in call_gtwy_agent - Added exception handler decorator to chat_multiple_agents function
- Added generate_test_cases bridge ID to bridge_ids and prebuilt_prompt_bridge_id constants
- Created new POST endpoint /{bridge_id}/generateAdditionalTestCases for generating test cases
- Implemented generate_additional_test_cases service to generate 10 test cases using AI based on system prompt
- Added parse_and_save_testcases function to process AI-generated test cases and save to database
- Handles both dict and list formats
Added default value of 0 for bridge_status when not found in bridge_data to prevent None comparison issues in status check
add api limit validations for folders
grok model response in dict formate
fix: add default value for bridge_status validation
…_default fix: set default bridge_status to 1 when creating new bridges
feat: add AI-powered test case generation endpoint with automatic saving
…are-python into testing
…vert parentDocId to string
feat: enhance get_bridges_with_tools_and_apikeys to conditionally con…
…on in process_variable_state
fix: remove unused var_state_value and simplify default value conditi…
…input fix: handle null input in anthropic tool calls and remove trailing wh…
…top-level response_format in process_background_tasks_for_playground
fix: use bridge_configurations playground_response_format instead of …
…on (#1094) The MongoDB session already handles transaction lifecycle automatically, so manual commit_transaction() and abort_transaction() calls are redundant and can be removed.
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.
No description provided.