feat: update embedding and LLM provider configurations#45
Merged
Conversation
- Simplified the ContextSeekMiddleware by delegating backend selection and embedder bridging to the canonical factory. - Updated the handling of LangChain embeddings and summarizers to prioritize user-supplied models, avoiding unnecessary instantiation of LLM instances. - Removed unused imports and redundant code to enhance clarity and maintainability.
- Introduced `auto_dream` feature to ContextSeekMiddleware, allowing for periodic dreaming based on a dual-gate mechanism: a count threshold (`dream_every`) and a minimum time interval since the last dream (`dream_min_interval_seconds`). - Updated `after_agent` method to trigger dreaming alongside compacting, ensuring both processes are managed independently. - Added tests to validate the new dreaming behavior, including scenarios for count and time gating, and ensuring dreaming does not interfere with compacting tasks.
…functionality - Expanded the documentation for `ContextSeekMiddleware` to detail the newly introduced `auto_dream` feature, including its parameters and behavior. - Added a section on the dual-gate mechanism for triggering dreams, explaining the conditions required for execution. - Included example code demonstrating the configuration of the `auto_dream` feature within the middleware setup.
- Bump version in package.json and package-lock.json to 0.2.0. - Introduce new chart components: BarList, DonutChart, LineChart, StatCards, and StatRows. - Add @radix-ui/react-popover dependency for enhanced UI components. - Refactor App component to include new panels and update navigation. - Implement theme and language context providers for improved user experience. - Enhance ScopeSelector with a dropdown for selecting scopes. - Update EmptyState and HealthBadge components to utilize i18n for localization.
- Removed IngressPanel from the App and navigation context. - Updated globalOverview API to accept an optional scope parameter. - Enhanced OverviewPanel to utilize the scope context for fetching data. - Refactored SettingsPanel to include additional configuration options for various storage backends. - Improved localization in i18n files with new settings and descriptions. - Cleaned up SkillsPanel by removing unused type detection logic and optimizing skill display logic.
- Added new configuration options for LLM and embedding providers, including base URLs and API keys. - Implemented update functionality for configuration via a new API endpoint. - Introduced editable rows in the SettingsPanel for user-friendly configuration management. - Enhanced localization support with new settings in i18n files for multiple languages. - Added install and restart functionalities to the context client for improved backend management. - Updated Vite configuration to include new API segments for installation and restart operations.
- Revised .env.example to clarify storage backend settings for seekdb and oceanbase modes. - Updated README and localization files to change "Browse" to "Insight" for better clarity in the dashboard UI. - Enhanced WritePanel to support tag merging from both input and JSON formats for improved data handling. - Improved documentation for CLI and storage guides to reflect the new storage backend configurations.
- Added a function to determine if the application is running as a desktop app, adjusting storage backend options accordingly in the SettingsPanel. - Implemented a polling mechanism in the Tauri frontend to check the backend status and update the UI based on health status. - Introduced a new command in the Rust backend to return the backend's health status, including the port and URL for access. - Ensured the desktop configuration file is created if it does not exist, setting default values for storage and model configurations. - Improved error handling in the environment file update process to create necessary directories if they do not exist.
- Revised .env.example to reflect updated provider options for embeddings and LLMs, including "openai", "dashscope", "ollama", and "huggingface". - Enhanced SettingsPanel to support new provider selections and added functionality for testing configurations. - Introduced new API endpoints for testing embedding and LLM connections, improving user experience in configuration management. - Updated localization files to include new settings and descriptions for better clarity across multiple languages.
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.
Summary
Related to #15
Related to #14