v0.8.0

Added

• LearningMachine-backed agent learning with culture updates and reasoning tools for synthesis agents
• Event-driven token usage tracking from model request events

Changed

• Upgraded Agno dependency to >=2.4.7
• BREAKING: Routing simplified to mandatory full_exploration (legacy single/double/triple strategy modes removed)
• Workflow now runs complexity analysis + full sequence only (condition-based simple/full branching removed)
• Forced strategy validation now rejects legacy strategy keys and accepts only full_exploration
• sequentialthinking tool contract now documents multi-step loop orchestration in tool description and input schema metadata
• sequentialthinking now publishes outputSchema and returns structuredContent control fields (should_continue, next_thought_number, stop_reason, and parameter guidance)
• Tool description and usage guidance now explicitly require active reflection (isRevision=true when correcting prior steps)

Fixed

• Fixed type error in thinking sequence synthesis in MultiThinkingSequentialProcessor

v0.7.0 - Parallel Execution and Enhanced Multi-Thinking

Major Features

🚀 Parallel Execution Architecture

• All thinking agents (except synthesis) now run asynchronously using asyncio.gather
• Significant performance improvement for multi-agent processing
• Intelligent orchestration between parallel and synthesis stages

🤖 Enhanced Multi-Thinking System

• Updated Groq provider with OpenAI GPT-OSS models (20B standard, 120B enhanced)
• Improved agent coordination and response synthesis
• Better handling of complex philosophical and creative thinking

🛠️ Infrastructure Improvements

• Fixed critical MetricsLogger import error that blocked server startup
• Streamlined logging system with structured output
• Enhanced error handling and performance tracking

📚 Documentation Overhaul

• Complete README restructure with detailed architecture diagrams
• Added comprehensive Mermaid flowcharts showing parallel execution
• Bilingual documentation (English/Chinese) with consistent formatting
• Removed all emoji characters for cleaner presentation

Technical Changes

• MetricsLogger implementation in logging infrastructure
• Parallel processing in multi_thinking_processor.py
• Groq provider model configuration updates
• Documentation syntax fixes and structural improvements

Compatibility

• Requires Python 3.10+
• Agno v2.0.5+ framework integration
• MCP (Model Context Protocol) server architecture

v0.5.0 - Major Agno v2.0 Migration Release

🚀 Major Release: Agno v2.0 Migration & Architectural Improvements

This release marks a significant milestone with the migration to Agno v2.0 and comprehensive architectural improvements.

🔥 Breaking Changes

• Agno v2.0 Migration: Complete migration with ~10,000x faster agent creation and ~50x less memory usage
• Src Layout: Adopted modern Python project structure following PEP 621 standards

✨ New Features

• Comprehensive TDD Coverage: Enhanced test infrastructure with 290+ test cases
• Adaptive Architecture: Cost optimization and intelligent routing capabilities
• Enhanced Constants: Extracted magic numbers for better maintainability
• Advanced Type Safety: Modernized codebase with enhanced type annotations

🏗️ Architecture Updates

• Team Coordination: Updated to Agno v2.0 coordination attributes
• Tool Migration: Migrated from ThinkingTools to ReasoningTools
• Memory Management: Updated to enable_user_memories parameter
• Performance: Microsecond-level factory initialization

🐛 Bug Fixes

• Resolved mypy type checking errors across all modules
• Comprehensive security and quality improvements
• Updated minimum Agno version to 2.0.5 for compatibility

📚 Documentation

• Updated CLAUDE.md with Agno v2.0 migration details
• Enhanced guidance for src layout and development requirements
• Improved test documentation and GitHub provider information

🔧 Technical Details

• Minimum Requirements: Python 3.10+, Agno ≥2.0.5
• Performance: ~10,000x faster agent creation, ~50x less memory usage
• Compatibility: Backward compatible APIs, same environment variables
• Testing: Comprehensive async test configuration with coverage reporting

This release maintains full backward compatibility while delivering massive performance improvements and enhanced code quality through Test-Driven Development practices.

---

🤖 Generated with Claude Code