-
Notifications
You must be signed in to change notification settings - Fork 645
RFC: [Go] Session snapshots #4232
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Summary of ChangesHello @apascal07, 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 proposes a detailed design for implementing session snapshots within the Genkit Go framework. The primary goal is to provide robust mechanisms for capturing, storing, and restoring the state of conversational agents, which will significantly improve debugging capabilities, allow users to resume conversations from any point, and enhance the observability of agent interactions through the developer UI. The design covers API definitions, behavioral specifications, usage examples, and implementation considerations, ensuring a flexible and extensible system for managing agent state history. Highlights
🧠 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 AssistThe 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
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 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
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Code Review
This design document provides a comprehensive and well-structured overview of the session snapshots feature in Genkit Go. The API definitions, behavioral descriptions, and usage examples are clear and helpful. The rationale section effectively explains the design decisions. I have a few minor suggestions for clarification regarding the TurnIndex definition and the computeStateHash function, which are detailed in the specific review comments.
| Index int `json:"index"` | ||
|
|
||
| // TurnIndex is the turn number when this snapshot was created (0-indexed). | ||
| // Turn 0 is after the first user input and agent response. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The definition of TurnIndex as "Turn 0 is after the first user input and agent response" might be slightly ambiguous. If Turn 0 is after the first full turn, it implies the initial state before any interaction would not have a TurnIndex of 0. Consider clarifying if TurnIndex is 0-indexed for the state before any user input, or if it represents the number of completed turns. For example, if TurnIndex is 0 for the initial state, then after the first user input and agent response, it would become 1.
| return true | ||
| } | ||
| // Compare by computing content-addressed IDs | ||
| return computeStateHash(snap.State) != computeStateHash(*snap.PrevState) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The computeStateHash function is used here to compare states but is not defined or explained anywhere in the document. Please add a brief explanation of what this function does (e.g., "a helper function that computes a SHA256 hash of the JSON-serialized state") or provide a reference to its definition if it exists elsewhere.
| | `SnapshotEventInvocationEnd` | When the agent function returns | | ||
|
|
||
| At each point, the snapshot callback is invoked. If it returns true: | ||
| 1. Compute the snapshot ID by hashing the state (SHA256) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No description provided.