docs: clarify Auto Memory proposes memory updates and skills#26527
docs: clarify Auto Memory proposes memory updates and skills#26527SandyTao520 merged 1 commit intomainfrom
Conversation
Summary of ChangesHello, 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 updates the Auto Memory documentation to accurately reflect its expanded capabilities. The changes clarify that the feature now supports both memory patches and skill drafts, provides updated guidance on the review process, and includes important details regarding data privacy and system prerequisites. These updates ensure users have a clear understanding of how the system proposes, validates, and applies changes to their environment. Highlights
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 the 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 counterproductive. 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. Footnotes
|
There was a problem hiding this comment.
Code Review
This pull request updates the documentation for the experimental Auto Memory feature to reflect its expanded capabilities. Beyond extracting reusable Agent Skills, the feature now supports proposing durable memory updates via unified diff .patch files. The documentation has been revised to include details on the new review process in the /memory inbox, safety boundaries that prevent direct file edits, and updated prerequisites for session analysis. I have no feedback to provide as there were no review comments.
Note: Security Review has been skipped due to the limited scope of the PR.
|
Size Change: -4 B (0%) Total Size: 34 MB
ℹ️ View Unchanged
|
Summary
Updates the Auto Memory documentation to reflect that the feature now proposes
both durable memory updates (as
.patchfiles) and reusable AgentSkills, rather than skills alone. Clarifies the review/approval flow, the
safety boundaries on what Auto Memory can and cannot touch, and the data/privacy
implications of model-based extraction.
Details
describe both memory patches and skill drafts.
the apply/dismiss actions for private vs. global memory patches.
sessions ignored) and the throttling/state-file behavior.
GEMINI.mdfiles are out of scope for auto-extraction— only private project memory, global personal memory, and skills can be
proposed.
be sent to the configured model during extraction.
docs/cli/tutorials/memory-management.md.No code changes; documentation only.
Related Issues
Related to #18007
How to Validate
docs/cli/auto-memory.mdand verify the rendered Markdown readscleanly (headings, lists, callouts, links).
./skills.mdand./skills.md#skill-discovery-tiers../tools/memory.md./settings.md#experimental./tutorials/memory-management.mddocs/cli/tutorials/memory-management.mdand confirm the updatedAuto Memory cross-link wording.
npm run lintto confirm no Markdown/Prettier regressions.Pre-Merge Checklist