Conversation
The automation configuration has been moved from TOML files to code-level configuration through the AutomationConfig struct. Updated documentation to reflect the new configuration approach and added parameters for max_concurrent_llm_tasks and generate_layers_every_n_messages.
The `log` crate has been removed from core modules and all logging calls have been updated to use the `tracing` crate instead, with appropriate log level adjustments.
Supports workspace-managed versions in Cargo.toml and adds --auto-confirm flag. Refactors version checking to verify specific versions are published.
Switch from using curl commands to native Node.js HTTPS module for checking crate availability on crates.io. This removes the dependency on curl and improves cross-platform support, especially on Windows.
capabilities The changes introduce entity preservation in abstract generation by adding a known_entities parameter to the generate_with_llm method. This ensures named entities are retained verbatim in generated abstracts. The abstract generation prompt was updated to handle entity preservation, and all callers were updated to pass an empty entity list by default. Additionally, the search functionality was enhanced with intent-based query analysis that supports multiple language types and query patterns. A new unified query analysis prompt was added that returns rewritten queries, keywords, entities, intent types, and time constraints in a single LLM call. The embedding client now includes built-in rate limiting and caching to improve API usage efficiency. Several configuration structs were updated to include cache and rate limiting parameters, and the embedding cache module was removed in favor of the new built-in caching.
Replace the old asynchronous AutoExtractor with MemoryEventCoordinator for synchronous memory extraction during session close. This provides better guarantees that memory processing completes before the CLI exits.
Introduce MemoryEventCoordinator for unified event processing and ensure stable content hashing with SHA-256. Enhance TARS UI with better error notifications.
language rules for memory extraction Ensure L0/L1 files are generated synchronously after memory extraction in on_session_closed to prevent missed updates on process exit. Also add critical language rules to preserve original language and technical terms in memory extraction.
Migrate from env_logger to tracing for better structured logging support. The new setup uses tracing-subscriber with a custom FileLayer that writes logs to both file and memory buffer, while maintaining backward compatibility through tracing-log bridge for existing log:: macros.
Skip timeline layer generation when L0/L1 files already exist to prevent redundant operations. Also optimize vector database indexing to only process current session files instead of all historical data.
cascade layer updates
Prevent duplicate layer cascade updates when session closure triggers update_all_layers simultaneously with background event processing of MemoryCreated/MemoryUpdated events.
- Change tool definitions from inputSchema/handler to parameters/execute - Update PluginAPI interface to match OpenClaw's pluginConfig - Make config properties optional with default values - Remove unused entry point and tools array from plugin manifest
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.
No description provided.