Packages
Structured causal reasoning chain logging for LLMs. Captures decision-making processes with cryptographic verification for transparency and debugging.
Current section
Files
Jump to
Current section
Files
crucible_trace
CHANGELOG.md
CHANGELOG.md
# Changelog
All notable changes to this project will be documented in this file.
## [0.1.0] - 2025-10-07
### Added
- Initial release
- Structured causal reasoning chain logging for LLM code generation
- Event tracking with six event types (hypothesis_formed, alternative_rejected, constraint_evaluated, pattern_applied, ambiguity_flagged, confidence_updated)
- LLM integration with XML-based event parsing
- Chain management for organizing reasoning events
- Persistent storage with JSON format and search capabilities
- Interactive HTML visualization with filtering and statistics
- Analysis tools for querying events, calculating statistics, and finding decision points
- Multiple export formats (JSON, Markdown, CSV)
### Documentation
- Comprehensive README with examples
- API documentation for all modules
- Usage examples for LLM integration and debugging
- Best practices for transparency in AI code generation