apecloud/ApeRAG

AI Summary: Modify the ApeRAG `list_collections` function to return only essential information (title, description, ID) for LLM search, removing sensitive data for security and efficiency.

AI Summary: Stabilize the existing unit tests for the ApeRAG project to improve the reliability of the test suite and facilitate a test-driven development approach. This involves identifying and fixing the root causes of frequent test failures in the current unit test suite.

AI Summary: Implement user quota limits for ApeRAG, including configurable limits on the number of collections, documents per collection, bots, and API keys. This requires modifications to the backend to track and enforce these limits, as well as updates to the UI to display and manage them.

AI Summary: Implement a user-friendly configuration system for ApeRAG to allow users to select from a curated list of LLMs optimized for different scenarios (agentic tasks and graph indexing). This involves defining default LLMs for each scenario and providing an easy-to-use interface for both SaaS and self-hosted deployments.

AI Summary: Fix a concurrency issue in the `/collections/{collection_id}/graphs/merge-suggestions` API endpoint of the ApeRAG project. The current implementation produces duplicate recommendation results when multiple requests are made concurrently. The solution requires ensuring that only unique results are returned, even under high concurrency.

AI Summary: Implement a 'clear chat history' feature in the ApeRAG frontend. This involves adding a button or command that prevents past messages from being sent to the bot for subsequent interactions while retaining the displayed chat history for user reference. This requires modifying the frontend (React/TypeScript) to manage message sending logic and potentially updating the backend (FastAPI) to handle the new command if necessary.

AI Summary: Implement an automatic chat naming feature for the ApeRAG platform. This involves analyzing the initial (and potentially subsequent) messages of a chat session to generate a concise and descriptive name. The feature should be user-configurable (on/off) and allow manual editing of the automatically generated names. This requires integration with the existing ApeRAG frontend and backend, potentially leveraging NLP techniques for content analysis.