memory: improving knowledge extraction, memory service integration, and RPC option handling - #3043
Merged
Merged
Conversation
Migrates browser knowledge/search/indexing from the legacy website collection to the durable memory service, adding RPC-safe service facades, corpus clearing, KnowPro-backed indexing, and memory-backed graph/analytics queries. Also improves extension import progress routing and offscreen download cancellation/retry handling.
Ask-about-page now requests answer generation for the current URL, and durable web search can synthesize grounded answers from indexed page evidence. Import progress forwarding now uses the active client event helper with debug logging, and temporary browser memory acceptance artifacts were removed.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This pull request introduces several updates across the agent server and browser agent packages, focusing on improving knowledge extraction, memory service integration, and RPC option handling. The most significant changes include enhanced support for nested RPC option methods, a more robust memory service initialization using a new index factory, and improvements to the browser agent’s knowledge extraction handler. Additionally, various JSON state files were added to support website import tracking.
Agent RPC and Knowledge Extraction Improvements
Support for Nested Option Methods in RPC: The RPC client now correctly resolves and invokes deeply nested option methods with the proper
thiscontext, and throws a clear error if the callback is not a function. A new test was added to verify this behavior. [1] [2]Improved Knowledge Extraction Handler: The browser agent’s
extractKnowledgeFromPagehandler has been refactored to support both direct HTML fragments and dynamic extraction via browser control, with sensible defaults for entity and relationship extraction. [1] [2]Memory Service and Indexing
Custom Index Factory for Memory Service: The agent server now initializes the
FileMemoryServicewith a custom index factory that creates a KnowPro corpus index using OpenAI’s GPT-5.6 Luna model and document memory settings. This enables more advanced knowledge indexing. [1] [2]Memory Service RPC Facade: The browser agent now uses a memory service RPC facade instead of an in-process client, improving modularity and potentially supporting remote memory service access.
Maintenance and Cleanup
Miscellaneous
Package Export Cleanup: The browser agent’s
package.jsonexports were simplified, removing theindexingexport.Refactored Imports: Updated imports in browser agent files to reflect new module structure and remove unused dependencies. [1] [2] [3]
These changes collectively enhance the modularity, reliability, and extensibility of the agent platform, especially regarding knowledge extraction and memory management.