This release transforms the deep research system from iterative editing to a section-based
workflow with extensive todo tracking and introduces auto-accept functionality for diff prompts.

Major Features:
- Section-based deep research workflow with single summarizer handling all sections sequentially
- Auto-accept functionality for diff prompts ('a'/'all' option)
- Comprehensive todo integration throughout all research phases
- 3-round iterative refinement per section (optimized from 5 rounds)
- Target 1-2 pages (500-1000 words) per section with executive-level writing
- Cat command concatenation for clean final assembly

Key Improvements:
- More efficient research workflow with better task visibility
- Reduced manual intervention through auto-accept diff functionality
- Professional quality output with iterative section refinement
- Complete todo tracking for research progress transparency

Package available on PyPI: https://pypi.org/project/mcp-cli-agent/1.2.4/