Fix/memory

[suluyana]

Change Summary

Related issue number

Checklist

• The pull request title is a good summary of the changes - it will be used in the changelog
• Unit tests for the changes exist
• Run pre-commit install and pre-commit run --all-files before git commit, and passed lint check.
• Documentation reflects the changes where applicable

[gemini-code-assist]

Summary of Changes

Hello @suluyana, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request primarily focuses on enhancing the memory management system and optimizing resource utilization within the agent. Key improvements include implementing asynchronous memory saving to prevent blocking operations, refining how messages are processed and cached in memory, and introducing lazy loading for the aiohttp library in various tools to reduce initial overhead. These changes aim to improve the agent's performance, responsiveness, and overall efficiency.

Highlights

• Asynchronous Memory Saving: Memory addition after a task now runs in a separate thread using asyncio.run_in_executor to prevent blocking the main event loop, improving responsiveness.
• Refined Memory Message Processing: A new parse_messages method has been introduced in default_memory to filter messages based on roles before they are added to memory, alongside improved cache analysis logic for more accurate message handling.
• Lazy Loading of aiohttp: The aiohttp library is now imported only when needed within specific functions in image and video generation tools, optimizing initial load time and resource usage.
• Flexible Fact Extraction Prompt Configuration: The configuration for the custom fact extraction prompt now allows for more flexible overriding and ensures the current date is only appended if a prompt is explicitly defined.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces several changes to improve memory management within the agent. Key changes include refactoring the add_memory function to be more explicit about when memory is added, fixing a bug in the message cache analysis logic, and moving some imports to be local to reduce initial load time and make dependencies optional.

My review focuses on two areas for improvement:

1. In llm_agent.py, the method for running an async memory operation in the background can be greatly simplified by using asyncio.create_task instead of the current run_in_executor with asyncio.run pattern.
2. In default_memory.py, there's a redundant check that can be removed to make the code cleaner.

Overall, the changes are positive and improve the robustness of the memory system. Addressing the suggested points will further enhance code quality and maintainability.

[gemini-code-assist]

The current approach to run add_memory as a background task using loop.run_in_executor combined with asyncio.run is unnecessarily complex and can be inefficient. It creates a new event loop in a separate thread for the coroutine.

A more idiomatic and efficient way to achieve this in an async function is to use asyncio.create_task. This schedules the coroutine to run on the current event loop without blocking the main flow.

            asyncio.create_task(self.add_memory(messages, add_type='add_after_task', **kwargs))

[gemini-code-assist]

The search method, which is called by this function, already handles the case where meta_infos is None. This check is therefore redundant and can be removed to simplify the code.