diff --git a/.claude/workflows/python-tools-strategy.md b/.claude/workflows/python-tools-strategy.md
deleted file mode 100644
index 636aa399..00000000
--- a/.claude/workflows/python-tools-strategy.md
+++ /dev/null
@@ -1,223 +0,0 @@
----
-name: python-tools-strategy
-description: Strategic framework for Python-based intelligent tool selection
-type: strategic-guideline
----
-
-# Python Tools Selection Strategy
-
-## ⚡ Core Framework
-
-**Python CLI (pycli)**: Unified interface for intelligent context discovery and tool execution
-**Vector Database**: Hierarchical semantic search with automatic parent discovery
-**Smart Analysis**: Context-aware file selection with similarity scoring
-
-### Decision Principles
-- **Context first** - Use vector search for intelligent file discovery
-- **Hierarchical by default** - Leverage parent directory vector databases automatically
-- **Semantic over syntactic** - Find relevant code by meaning, not just keywords
-- **Tool integration** - Seamlessly combine context discovery with Gemini/Codex execution
-
-### Quick Decision Rules
-1. **Need context discovery?** → Start with `pycli --analyze --query`
-2. **Know exact files?** → Use `pycli --analyze -p` directly
-3. **First time in project?** → Run `pycli --init` first
-4. **Files changed?** → Update with `pycli --update-embeddings`
-
-### Core Execution Rules
-- **Default Tool**: Gemini for analysis, Codex for development
-- **Similarity Threshold**: 0.3 minimum for relevant results
-- **Hierarchical Search**: Automatic parent directory vector database discovery
-- **Command Pattern**: Always use `pycli` wrapper for consistent interface
-
-## 🎯 Universal Command Template
-
-### Standard Format (REQUIRED)
-```bash
-# Smart Context Discovery
-pycli --analyze --query "
-PURPOSE: [clear analysis goal]
-SEARCH: [semantic search terms]
-TOOL: [gemini/codex/both]
-EXPECTED: [expected context and output]
-" --tool [gemini/codex]
-
-# Direct Tool Execution
-pycli --analyze --tool [gemini/codex] -p "
-PURPOSE: [clear execution goal]
-TASK: [specific execution task]
-CONTEXT: [known file references]
-EXPECTED: [expected deliverables]
-"
-```
-
-### Template Structure
-- [ ] **PURPOSE** - Clear goal and intent for analysis
-- [ ] **SEARCH/TASK** - Semantic search terms or specific task
-- [ ] **TOOL** - Gemini for analysis, Codex for development
-- [ ] **CONTEXT** - File references and project context
-- [ ] **EXPECTED** - Clear expected results and format
-
-## 📊 Tool Selection Matrix
-
-| Task Type | Command | Use Case | Context Strategy |
-|-----------|---------|----------|------------------|
-| **Context Discovery** | `pycli --analyze --query` | Code exploration, pattern finding | Vector similarity search |
-| **Targeted Analysis** | `pycli --analyze --tool gemini -p` | Architecture review, understanding | Known file analysis |
-| **Development** | `pycli --analyze --tool codex -p` | Feature implementation, bug fixes | Smart context + execution |
-| **Setup** | `pycli --init` | Project initialization | Vector database creation |
-| **Maintenance** | `pycli --update-embeddings` | Index updates after changes | Incremental vectorization |
-| **Health Check** | `pycli --status` | System verification | Database validation |
-
-## 🚀 Usage Patterns
-
-### Workflow Integration (REQUIRED)
-When planning any coding task, **ALWAYS** integrate Python CLI tools:
-
-1. **Discovery Phase**: Use `pycli --analyze --query` for context
-2. **Analysis Phase**: Use Gemini for understanding with smart context
-3. **Implementation Phase**: Use Codex for development with relevant files
-4. **Validation Phase**: Update embeddings and verify results
-
-### Common Scenarios
-```bash
-# Project Context Discovery
-pycli --analyze --query "
-PURPOSE: Understand authentication architecture
-SEARCH: authentication patterns, login systems, user management
-TOOL: gemini
-EXPECTED: Architecture overview and key implementation files
-" --tool gemini
-
-# Feature Development with Context
-pycli --analyze --query "
-PURPOSE: Implement user registration
-SEARCH: user creation, validation patterns, database models
-TOOL: codex
-EXPECTED: Complete registration module with tests
-" --tool codex
-
-# Code Quality Analysis
-pycli --analyze --query "
-PURPOSE: Review error handling patterns
-SEARCH: exception handling, error middleware, logging
-TOOL: gemini
-EXPECTED: Error handling assessment and recommendations
-" --tool gemini
-```
-
-## 📋 Planning Checklist
-
-For every development task:
-- [ ] **Discovery completed** - Context discovery with vector search
-- [ ] **Purpose defined** - Clear goal and intent documented
-- [ ] **Tool selected** - Gemini for analysis, Codex for development
-- [ ] **Context gathered** - Relevant files identified through similarity
-- [ ] **Template applied** - Standard command format used
-- [ ] **Embeddings updated** - Vector database reflects current state
-- [ ] **Results validated** - Output quality and relevance verified
-
-## 🎯 Key Features
-
-### Python CLI (pycli)
-- **Command**: `pycli --analyze`
-- **Strengths**: Hierarchical vector search, semantic similarity, context discovery
-- **Best For**: Intelligent file selection, context-aware analysis, project exploration
-
-### Vector Database
-- **Hierarchical**: Automatic parent directory discovery
-- **Semantic**: Meaning-based similarity scoring
-- **Efficient**: Incremental updates and smart caching
-- **Scalable**: Project-wide context with subdirectory support
-
-### Context Patterns
-- Query-based: `--query "semantic search terms"`
-- Direct prompt: `-p "specific task"`
-- Tool selection: `--tool [gemini|codex|both]`
-- Similarity control: `--top-k N --similarity-threshold X`
-
-## 🔧 Best Practices
-
-- **Start with discovery** - Use `--query` for context before direct prompts
-- **Be semantic** - Use meaning-based search terms, not just keywords
-- **Update regularly** - Run `--update-embeddings` after file changes
-- **Validate context** - Check similarity scores and relevance before proceeding
-- **Document patterns** - Reference successful query patterns for reuse
-- **Leverage hierarchy** - Work in subdirectories, let parent DBs provide context
-
-## 📁 Hierarchical Vector System
-
-**Base Structure**: `~/.claude/vector_db/[project-path]/`
-
-### Automatic Discovery
-```
-Project Structure                Vector Database Usage
-/project/                       Creates: ~/.claude/vector_db/project/
-├── src/                        → Uses parent DB automatically
-│   ├── auth/                   → Uses parent DB automatically
-│   └── api/                    → Uses parent DB automatically
-└── tests/                      → Uses parent DB automatically
-```
-
-### Smart Context Integration
-- **Parent Discovery**: Subdirectories automatically use parent vector DB
-- **Semantic Search**: Find files by meaning, not just filename patterns
-- **Similarity Scoring**: Relevance-based file selection with configurable thresholds
-- **Incremental Updates**: Efficient re-indexing of only changed files
-
-### Migration Benefits
-```bash
-# Enhanced Context Discovery (vs traditional grep/find)
-# OLD: find . -name "*auth*" | head -10
-# NEW: pycli --analyze --query "authentication patterns" --tool gemini
-
-# OLD: grep -r "login" src/ | head -20
-# NEW: pycli --analyze --query "login implementation" --tool codex
-
-# OLD: ~/.claude/scripts/gemini-wrapper -p "analyze auth"
-# NEW: pycli --analyze --query "authentication architecture" --tool gemini
-```
-
-## 🎯 Quick Setup Guide
-
-
-### Project Setup (One-time per project)
-```bash
-# 1. Navigate to project root
-cd /path/to/project
-
-# 2. Initialize vector database
-pycli --init
-
-# 3. Verify setup
-pycli --status
-```
-
-### Daily Workflow
-```bash
-# 1. Update embeddings (after file changes)
-pycli --update-embeddings
-
-# 2. Smart context discovery
-pycli --analyze --query "your search terms" --tool gemini
-
-# 3. Targeted development
-pycli --analyze --query "implementation patterns" --tool codex
-```
-
-## 🎪 Decision Framework
-
-```
-Need intelligent code analysis?
-├─ Know specific files to analyze?
-│  ├─ YES → pycli --analyze --tool [gemini/codex] -p "prompt"
-│  └─ NO → pycli --analyze --query "semantic search" --tool [gemini/codex]
-├─ Vector database updated?
-│  ├─ UNSURE → pycli --status
-│  ├─ NO → pycli --update-embeddings
-│  └─ YES → Proceed with analysis
-└─ First time in project?
-   └─ Run pycli --init first
-```
-
-
diff --git a/README.md b/README.md
index 12c99ca6..e34e3259 100644
--- a/README.md
+++ b/README.md
@@ -16,12 +16,12 @@
 
 **Claude Code Workflow (CCW)** is a next-generation multi-agent automation framework for software development that orchestrates complex development tasks through intelligent workflow management and autonomous execution.
 
-> **🎯 Latest Release v2.0**: Major architectural evolution with enhanced workflow lifecycle, Python-powered backend (`pycli`), comprehensive test workflow generation, plan verification system, and brainstorm artifacts integration. See [CHANGELOG.md](CHANGELOG.md) for details.
+> **🎯 Latest Release v2.0**: Major architectural evolution with enhanced workflow lifecycle, comprehensive test workflow generation, plan verification system, and brainstorm artifacts integration. See [CHANGELOG.md](CHANGELOG.md) for details.
 
 ### 🌟 Key Innovations
 
 - **🔄 Enhanced Workflow Lifecycle**: Complete development cycle: Brainstorm → Plan → Verify → Execute → Test → Review
-- **🐍 Python-Powered Backend**: Advanced `pycli` integration with hierarchical vector database support for semantic context discovery
+- **🧪 Automated Test Generation**: Comprehensive test workflow creation with full coverage planning
 - **🧪 Automated Test Generation**: Comprehensive test workflow generation (`/workflow:test-gen`) with full coverage planning
 - **✅ Plan Verification System**: Pre-execution validation using dual Gemini/Codex analysis (`/workflow:plan-verify`)
 - **🎯 JSON-First Architecture**: Single source of truth with atomic session management
@@ -105,19 +105,12 @@ Complete development lifecycle with quality gates at each phase:
 5. **🧪 Test Phase** - Automated test workflow generation with comprehensive coverage
 6. **🔍 Review Phase** - Quality assurance and completion validation
 
-### 🐍 **Python-Powered Backend (`pycli`)**
-Advanced semantic context discovery system:
-
-```bash
-# Hierarchical vector database support
-pycli --semantic-search "authentication patterns"
-
-# Auto-detects parent directory vector DB
-pycli --context-discovery --inherit-parent
-
-# Intelligent file relevance scoring
-pycli --analyze-relevance "jwt token validation"
-```
+### 🧪 **Automated Test Generation**
+Comprehensive test workflow creation:
+- **Implementation Analysis**: Scans completed IMPL-* tasks for test requirements
+- **Multi-layered Testing**: Unit, Integration, E2E, Performance, Security tests
+- **Agent Assignment**: Specialized test agents for different test types
+- **Dependency Mapping**: Test execution follows implementation dependency chains
 
 ### 🧪 **Automated Test Generation**
 Comprehensive test workflow creation:
@@ -345,14 +338,13 @@ graph LR
 ├── 💬 prompt-templates/      # AI interaction templates
 ├── 🔧 scripts/              # Automation utilities
 │   ├── 📊 gemini-wrapper           # Intelligent Gemini wrapper
-│   ├── 🐍 pycli                    # NEW: Python CLI backend
-│   ├── 🛠️ install_pycli.sh          # NEW: Python tools installer
+│   ├── 🔧 get_modules_by_depth.sh  # Project analysis
 │   ├── 📋 read-task-paths.sh       # Task path conversion
 │   └── 🏗️ get_modules_by_depth.sh  # Project analysis
 ├── 🛠️ workflows/            # Core workflow documentation
 │   ├── 🏛️ workflow-architecture.md      # System architecture
 │   ├── 📊 intelligent-tools-strategy.md # Tool selection guide
-│   ├── 🐍 python-tools-strategy.md     # NEW: Python backend strategy
+│   ├── 🔧 context-search-strategy.md  # Search and discovery strategy
 │   └── 🔧 tools-implementation-guide.md # Implementation details
 └── ⚙️ settings.local.json   # Local configuration
 
@@ -394,7 +386,6 @@ graph LR
 - **🔮 Qwen CLI**: Required for architecture analysis and code generation
 - **📂 Git Repository**: Required for change tracking and version control
 - **🎯 Claude Code IDE**: Recommended for optimal experience
-- **🐍 Python 3.8+**: Required for advanced pycli backend features
 
 ---
 
diff --git a/README_CN.md b/README_CN.md
index 4d5e73ef..d3f29f2c 100644
--- a/README_CN.md
+++ b/README_CN.md
@@ -475,7 +475,6 @@ cd src/api && /update-memory-related
 - **🔮 Qwen CLI**: 架构分析和代码生成必需
 - **📂 Git仓库**: 变更跟踪和版本控制必需
 - **🎯 Claude Code IDE**: 推荐用于最佳体验
-- **🐍 Python 3.8+**: 高级pycli后端功能必需
 
 ## 配置