v0.1.11 - CLI Workflows & Streamlined Architecture

Added

• CLI Orchestrator for automated research workflows
  • aponyx run command for executing complete signal-strategy pipelines
  • aponyx report command for generating comprehensive analysis reports (console/markdown/HTML)
  • aponyx list command for browsing available signals, strategies, and datasets
  • aponyx clean command for managing cached workflow results with improved feedback
  • YAML configuration file support for reproducible workflows
  • Smart caching with automatic skip of completed steps
  • Force re-run option for invalidating cache
  • Subset execution for running specific workflow steps
  • Example workflow YAML files in examples/ directory
  • Comprehensive CLI user guide (src/aponyx/docs/cli_user_guide.md)
  • Help text shorthand (-h) for all commands
  • Comprehensive CLI layer test suite with 28 tests
• Workflow Engine infrastructure
  • WorkflowEngine for sequential pipeline execution with dependency tracking
  • WorkflowConfig for immutable workflow configuration
  • Protocol-based step abstraction with WorkflowStep interface
  • Six concrete workflow steps: data, signal, suitability, backtest, performance, visualization
  • StepRegistry for centralized step factory and ordering
  • Error handling with partial result preservation
  • Structured logging with progress indicators
  • Cache-then-raw fallback for workflow data loading
  • Decoupled data step from instrument-specific logic
• Reporting Package for analysis aggregation
  • generate_report() function with multiple output formats
  • Console reports with table formatting
  • Markdown reports with embedded visualization links
  • HTML reports with styled formatting
  • Automatic aggregation of suitability and performance evaluations
  • Smart report data collection from workflow outputs
• Example Scripts for standalone workflow execution
  • 01_generate_synthetic_data.py - Synthetic data generation
  • 02_fetch_data_file.py - File-based data loading
  • 03_fetch_data_bloomberg.py - Bloomberg Terminal data fetch
  • 04_compute_signal.py - Signal computation
  • 05_evaluate_suitability.py - Signal-product suitability assessment
  • 06_run_backtest.py - Strategy backtesting
  • 07_analyze_performance.py - Performance analysis
  • 08_visualize_results.py - Results visualization

Changed

• Architecture Improvements
  • Decoupled configuration from specific instruments for better modularity
  • Separated signal metadata, registry, and orchestration into distinct modules
  • Removed defensive error handling from example scripts for clarity
• CLI Enhancements
  • Simplified CLI output for better readability
  • Changed default product to cdx_ig_5y for consistency
  • Improved help formatting with better descriptions
• Documentation Overhaul
  • Consolidated and streamlined documentation structure
  • Updated all docs for CLI/workflow infrastructure
  • Fixed README Quick Start example and documentation links
  • Removed all notebook-related documentation
• Testing
  • Resolved all 28 failing tests
  • Added comprehensive CLI layer test coverage
• Dependencies
  • Updated pyproject.toml to include pyyaml>=6.0 dependency
  • Added aponyx CLI entry point in [project.scripts]

Removed

• Jupyter Notebooks - Complete removal of notebook workflow
  • Deleted entire src/aponyx/notebooks/ directory with all workflow notebooks
  • Removed NOTEBOOK_TROUBLESHOOTING.md
  • Removed notebook-related sections from README.md
  • Removed notebook context from copilot-instructions.md
  • Cleaned up notebook references from design docs (performance_evaluation, visualization, python_guidelines)
  • Removed notebook references from PROJECT_STATUS.md
  • CLI workflows now provide complete orchestration, making notebooks redundant