Skip to content
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

feat: Add AnnotationMessage type and ConversationMemory #241

Closed

Conversation

devin-ai-integration[bot]
Copy link
Contributor

@devin-ai-integration devin-ai-integration bot commented Nov 25, 2024

This PR adds two new features to PromptingTools.jl:

  1. AnnotationMessage type:
  • New concrete type that inherits from AbstractAnnotationMessage
  • Designed to store metadata alongside message history
  • Never sent to LLM APIs (filtered by render function)
  • Includes extras Dict, tags Vector{Symbol}, and comment String fields
  • Utility functions annotate! and annotate for easy usage
  1. ConversationMemory:
  • Efficient management of conversation history
  • Smart truncation with batch size awareness
  • System message and first user message preservation
  • Run ID-based deduplication
  • Direct integration with aigenerate

All tests are passing and the implementation follows the existing patterns in the codebase.

Link to Devin run: https://preview.devin.ai/devin/1313c322110e474eb7be51c54071a39c

If you have any feedback, you can leave comments in the PR and I'll address them in the app!

- Implement ConversationMemory struct for efficient message history management
- Add batch-aware truncation and caching capabilities
- Enhance AnnotationMessage with comprehensive filtering tests across providers
- Add tests for edge cases and multiple consecutive annotations
- Implement ConversationMemory with batch-aware message truncation
- Add AnnotationMessage type for metadata and documentation
- Add comprehensive test suites for both features
- Ensure proper rendering behavior for annotation messages
- Add AnnotationMessage type for metadata and documentation
- Implement ConversationMemory for efficient message history management
- Add comprehensive test suite for both features
- Update Project.toml dependencies
- Adjust test configurations
- Update test files for memory implementation
…tationMessage type for storing metadata with messages\n- Implement ConversationMemory for efficient message history management\n- Add comprehensive test suite for both features\n- Ensure annotation messages are filtered from LLM rendering
@svilupp svilupp closed this Nov 26, 2024
@svilupp svilupp deleted the devin/1732578965-add-annotation-messages branch November 26, 2024 10:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant