feat: Add experimental MCP tools integration for enhanced codebase analysis

## New Features - **MCP Tools Integration**: Added support for Model Context Protocol tools - Exa MCP Server: External API patterns and best practices - Code Index MCP: Advanced internal codebase search and indexing - **Enhanced Workflow Planning**: Updated pre_analysis to include MCP tool steps - **Documentation Updates**: Added MCP tool setup guides and usage examples ## Changes ### Core Components - Updated `plan.md` with MCP integration principles and implementation approach guidelines - Added MCP tool steps in pre_analysis workflow: `mcp_codebase_exploration`, `mcp_external_context` - Enhanced context accumulation with external best practices lookup ### Documentation - Added comprehensive MCP tools section in both English and Chinese README - Updated installation requirements and integration guidelines - Added GitHub repository links for required MCP servers ### Agent Enhancements - Updated multiple agents to support MCP tool integration - Enhanced context gathering capabilities with external pattern analysis ## Technical Details - MCP tools provide faster analysis through direct codebase indexing - Automatic fallback to traditional bash/CLI tools when MCP unavailable - Enhanced pattern recognition and similarity detection capabilities 🧪 **Experimental**: MCP integration is currently experimental and optional 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2026-02-05 01:50:27 +08:00 · 2025-09-28 16:40:01 +08:00
parent 075b4d1bbc
commit a4b32f23b8
15 changed files with 322 additions and 30 deletions
--- a/.claude/agents/action-planning-agent.md
+++ b/.claude/agents/action-planning-agent.md
@@ -71,6 +71,7 @@ You are a pure execution agent specialized in creating actionable implementation
 4. Use consolidated analysis to inform implementation stages and task breakdown

 #### Analysis Dimensions Coverage
+- **Exa Research**: Use `mcp__exa__get_code_context_exa` for technology stack selection and API patterns
 - Architecture patterns and component relationships
 - Implementation conventions and coding standards
 - Module dependencies and integration points
--- a/.claude/agents/code-developer.md
+++ b/.claude/agents/code-developer.md
@@ -85,6 +85,12 @@ ELIF context insufficient OR task has flow control marker:

 **Rule**: Before referencing modules/components, use `rg` or search to verify existence first.

+**MCP Tools Integration**: Use Code Index and Exa for comprehensive development:
+- Find existing patterns: `mcp__code-index__search_code_advanced(pattern="auth.*function")`
+- Locate files: `mcp__code-index__find_files(pattern="src/**/*.ts")`
+- Get API examples: `mcp__exa__get_code_context_exa(query="React authentication hooks", tokensNum="dynamic")`
+- Update after changes: `mcp__code-index__refresh_index()`
+
 **Test-Driven Development**:
 - Write tests first (red → green → refactor)
 - Focus on core functionality and edge cases
--- a/.claude/agents/code-review-test-agent.md
+++ b/.claude/agents/code-review-test-agent.md
@@ -159,6 +159,11 @@ if [FAST_MODE]: apply targeted review process
 - Missing or unused imports identification
 - Circular dependency detection

+**MCP Tools Integration**: Use Code Index for comprehensive analysis:
+- Pattern discovery: `mcp__code-index__search_code_advanced(pattern="import.*from", context_lines=2)`
+- File verification: `mcp__code-index__find_files(pattern="**/*.test.js")`
+- Post-review refresh: `mcp__code-index__refresh_index()`
+
 ### Performance
 - Algorithm complexity (time and space)
 - Database query optimization
--- a/.claude/agents/conceptual-planning-agent.md
+++ b/.claude/agents/conceptual-planning-agent.md
@@ -86,17 +86,17 @@ def handle_brainstorm_assignment(prompt):

 ### Role-Specific Analysis Dimensions

-| Role | Primary Dimensions | Focus Areas |
-|------|-------------------|--------------|
-| system-architect | architecture_patterns, scalability_analysis, integration_points | Technical design and system structure |
-| ui-designer | user_flow_patterns, component_reuse, design_system_compliance | UI/UX patterns and consistency |
-| business-analyst | process_optimization, cost_analysis, efficiency_metrics, workflow_patterns | Business process and ROI |
-| data-architect | data_models, flow_patterns, storage_optimization | Data structure and flow |
-| security-expert | vulnerability_assessment, threat_modeling, compliance_check | Security risks and compliance |
-| user-researcher | usage_patterns, pain_points, behavior_analysis | User behavior and needs |
-| product-manager | feature_alignment, market_fit, competitive_analysis | Product strategy and positioning |
-| innovation-lead | emerging_patterns, technology_trends, disruption_potential | Innovation opportunities |
-| feature-planner | implementation_complexity, dependency_mapping, risk_assessment | Development planning |
+| Role | Primary Dimensions | Focus Areas | Exa Usage |
+|------|-------------------|--------------|-----------|
+| system-architect | architecture_patterns, scalability_analysis, integration_points | Technical design and system structure | `mcp__exa__get_code_context_exa("microservices patterns")` |
+| ui-designer | user_flow_patterns, component_reuse, design_system_compliance | UI/UX patterns and consistency | `mcp__exa__get_code_context_exa("React design system patterns")` |
+| business-analyst | process_optimization, cost_analysis, efficiency_metrics, workflow_patterns | Business process and ROI | `mcp__exa__get_code_context_exa("business workflow automation")` |
+| data-architect | data_models, flow_patterns, storage_optimization | Data structure and flow | `mcp__exa__get_code_context_exa("database schema patterns")` |
+| security-expert | vulnerability_assessment, threat_modeling, compliance_check | Security risks and compliance | `mcp__exa__get_code_context_exa("security authentication patterns")` |
+| user-researcher | usage_patterns, pain_points, behavior_analysis | User behavior and needs | `mcp__exa__get_code_context_exa("user research methodologies")` |
+| product-manager | feature_alignment, market_fit, competitive_analysis | Product strategy and positioning | `mcp__exa__get_code_context_exa("product management frameworks")` |
+| innovation-lead | emerging_patterns, technology_trends, disruption_potential | Innovation opportunities | `mcp__exa__get_code_context_exa("emerging technology trends")` |
+| feature-planner | implementation_complexity, dependency_mapping, risk_assessment | Development planning | `mcp__exa__get_code_context_exa("agile development patterns")` |

 ### Output Integration

--- a/.claude/agents/general-purpose.md
+++ b/.claude/agents/general-purpose.md
@@ -31,6 +31,7 @@ You are a versatile execution specialist focused on completing high-quality task
 ### 1. Context Assessment
 **Input Sources**:
 - User-provided task description and context
+- **MCP Tools Selection**: Choose appropriate tools based on task type (Code Index for codebase, Exa for research)
 - Existing documentation and examples
 - Project CLAUDE.md standards
 - Domain-specific requirements
--- a/.claude/agents/memory-gemini-bridge.md
+++ b/.claude/agents/memory-gemini-bridge.md
@@ -15,11 +15,15 @@ Coordinate parallel execution of `~/.claude/scripts/update_module_claude.sh` scr

 ### 1. Analyze Project Structure
 ```bash
-# Step 1: Get module list with depth information
+# Step 1: Code Index architecture analysis
+mcp__code-index__search_code_advanced(pattern="class|function|interface", file_pattern="**/*.{ts,js,py}")
+mcp__code-index__find_files(pattern="**/*.{md,json,yaml,yml}")
+
+# Step 2: Get module list with depth information
 modules=$(Bash(~/.claude/scripts/get_modules_by_depth.sh list))
 count=$(echo "$modules" | wc -l)

-# Step 2: Display project structure
+# Step 3: Display project structure
 Bash(~/.claude/scripts/get_modules_by_depth.sh grouped)
 ```

--- a/.claude/commands/update-memory-full.md
+++ b/.claude/commands/update-memory-full.md
@@ -16,10 +16,14 @@ Complete project-wide documentation update using depth-parallel execution.
 #!/bin/bash
 # Complete project-wide CLAUDE.md documentation update

-# Step 1: Cache git changes
+# Step 1: Code Index architecture analysis
+mcp__code-index__search_code_advanced(pattern="class|function|interface", file_pattern="**/*.{ts,js,py}")
+mcp__code-index__find_files(pattern="**/*.{md,json,yaml,yml}")
+
+# Step 2: Cache git changes
 Bash(git add -A 2>/dev/null || true)

-# Step 2: Get and display project structure
+# Step 3: Get and display project structure
 modules=$(Bash(~/.claude/scripts/get_modules_by_depth.sh list))
 count=$(echo "$modules" | wc -l)

--- a/.claude/commands/update-memory-related.md
+++ b/.claude/commands/update-memory-related.md
@@ -17,10 +17,14 @@ Context-aware documentation update for modules affected by recent changes.
 #!/bin/bash
 # Context-aware CLAUDE.md documentation update

-# Step 1: Detect changed modules (before staging)
+# Step 1: Code Index refresh and architecture analysis
+mcp__code-index__refresh_index()
+mcp__code-index__search_code_advanced(pattern="class|function|interface", file_pattern="**/*.{ts,js,py}")
+
+# Step 2: Detect changed modules (before staging)
 changed=$(Bash(~/.claude/scripts/detect_changed_modules.sh list))

-# Step 2: Cache git changes (protect current state)
+# Step 3: Cache git changes (protect current state)
 Bash(git add -A 2>/dev/null || true)

 # Step 3: Use detected changes or fallback
--- a/.claude/commands/workflow/brainstorm/artifacts.md
+++ b/.claude/commands/workflow/brainstorm/artifacts.md
@@ -25,6 +25,7 @@ Dedicated command for topic discussion, decomposition, and analysis artifacts ge
 ### Discussion & Artifacts Generation Process

 **0. Session Management** ⚠️ FIRST STEP
+- **MCP Tools Integration**: Use Code Index for technical context, Exa for industry insights
 - **Active session detection**: Check `.workflow/.active-*` markers
 - **Session selection**: Prompt user if multiple active sessions found
 - **Auto-creation**: Create `WFS-[topic-slug]` only if no active session exists
--- a/.claude/commands/workflow/brainstorm/auto.md
+++ b/.claude/commands/workflow/brainstorm/auto.md
@@ -41,6 +41,7 @@ ls ~/.claude/workflows/cli-templates/planning-roles/  # Show all available roles
 The command performs dedicated role analysis through:

 **0. Session Management** ⚠️ FIRST STEP
+- **MCP Tools Integration**: Use Code Index for codebase context, Exa for external insights
 - **Active session detection**: Check `.workflow/.active-*` markers
 - **Session selection**: Prompt user if multiple active sessions found
 - **Auto-creation**: Create `WFS-[topic-slug]` only if no active session exists
--- a/.claude/commands/workflow/concept-eval.md
+++ b/.claude/commands/workflow/concept-eval.md
@@ -46,6 +46,7 @@ Pre-planning evaluation command that assesses concept feasibility, identifies po
 The command performs comprehensive concept evaluation through:

 **0. Context Preparation** ⚠️ FIRST STEP
+- **MCP Tools Integration**: Use Code Index for codebase exploration, Exa for external context
 - **Documentation loading**: Automatic context gathering based on concept scope
  - **Always check**: `CLAUDE.md`, `README.md` - Project context and conventions
  - **For architecture concepts**: `.workflow/docs/architecture/`, existing system patterns
--- a/.claude/commands/workflow/plan.md
+++ b/.claude/commands/workflow/plan.md
@@ -11,6 +11,24 @@ examples:

 # Workflow Plan Command

+## 🚀 MCP Tools Integration (NEW!)
+
+**Enhanced with MCP (Model Context Protocol) tools for advanced codebase analysis:**
+
+### Required MCP Servers
+1. **Exa MCP Server** - External API patterns and examples
+   - Repository: https://github.com/exa-labs/exa-mcp-server
+   - Function: `mcp__exa__get_code_context_exa()` - Get external best practices
+
+2. **Code Index MCP** - Internal codebase exploration
+   - Repository: https://github.com/johnhuang316/code-index-mcp
+   - Functions:
+     - `mcp__code-index__find_files()` - File pattern matching
+     - `mcp__code-index__search_code_advanced()` - Advanced code search
+
+### Installation & Setup
+Please install these MCP servers to enable enhanced codebase analysis. The workflow will automatically use them when available.
+
 ## Usage
 ```bash
 /workflow:plan <input>
@@ -91,6 +109,18 @@ The following pre_analysis steps are generated for agent execution:
      "command": "bash(cat .workflow/WFS-[session]/.process/ANALYSIS_RESULTS.md 2>/dev/null || echo 'planning analysis not found')",
      "output_to": "planning_context"
    },
+    {
+      "step": "mcp_codebase_exploration",
+      "action": "Explore codebase structure and patterns using MCP tools",
+      "command": "mcp__code-index__find_files(pattern=\"[task_focus_patterns]\") && mcp__code-index__search_code_advanced(pattern=\"[relevant_patterns]\", file_pattern=\"[target_extensions]\")",
+      "output_to": "codebase_structure"
+    },
+    {
+      "step": "mcp_external_context",
+      "action": "Get external API examples and best practices",
+      "command": "mcp__exa__get_code_context_exa(query=\"[task_technology] [task_patterns]\", tokensNum=\"dynamic\")",
+      "output_to": "external_context"
+    },
    {
      "step": "load_dependencies",
      "action": "Retrieve dependency task summaries",
@@ -104,18 +134,11 @@ The following pre_analysis steps are generated for agent execution:
      "output_to": "doc_context"
    },
    {
-      "step": "analyze_patterns",
-      "action": "Analyze codebase patterns and architecture using CLI tools with directory context",
-      "command": "bash(cd \"[target_directory]\" && ~/.claude/scripts/gemini-wrapper -p \"PURPOSE: Analyze existing patterns TASK: Identify implementation patterns for [task_type] CONTEXT: [planning_context] [dependency_context] EXPECTED: Pattern analysis and recommendations RULES: Focus on architectural consistency\")",
-      "output_to": "pattern_analysis",
-      "on_error": "skip_optional"
-    },
-    {
-      "step": "analyze_implementation",
-      "action": "Development-focused analysis using Codex when needed",
-      "command": "bash(codex -C [target_directory] --full-auto exec \"PURPOSE: Analyze implementation patterns TASK: Review development patterns for [task_type] CONTEXT: [planning_context] [dependency_context] EXPECTED: Development strategy and code patterns RULES: Focus on implementation consistency\" -s danger-full-access)",
-      "output_to": "implementation_analysis",
-      "on_error": "skip_optional"
+      "step": "gather_task_context",
+      "action": "Analyze task context and dependencies without implementation",
+      "command": "bash(cd \"[task_focus_paths]\" && ~/.claude/scripts/gemini-wrapper -p \"PURPOSE: Analyze task context and patterns TASK: Review existing patterns and dependencies for '[task_title]' CONTEXT: Task ID [task_id], Focus paths: [task_focus_paths], MCP findings: [codebase_structure] [external_context] EXPECTED: Pattern analysis, dependency mapping, and architectural insights RULES: Focus on understanding existing code patterns, no implementation\")",
+      "output_to": "task_context",
+      "on_error": "fail"
    }
  ]
 }
@@ -126,7 +149,49 @@ Flow_control design should follow these principles:
 1. **Structure Analysis**: Project hierarchy and patterns
 2. **Dependency Mapping**: Previous task summaries → inheritance context
 3. **Task Context Generation**: Combined analysis → task.context fields
-4. **CLI Tool Analysis**: Use Gemini/Codex appropriately for pattern analysis when needed 
+4. **CLI Tool Analysis**: Use Gemini/Codex appropriately for pattern analysis when needed
+
+**MCP Integration Principles**:
+- **Code Index First**: Use `mcp__code-index__` for internal codebase exploration before external tools
+- **Exa for Context**: Use `mcp__exa__get_code_context_exa` to supplement with external API patterns and examples
+- **Automatic Fallback**: If MCP tools unavailable, workflow uses traditional bash/CLI tools
+- **Enhanced Analysis**: MCP tools provide deeper codebase understanding and external best practices
+
+**Benefits of MCP Integration**:
+- **Faster Analysis**: Direct codebase indexing vs. manual file searching
+- **External Context**: Real-world API patterns and implementation examples
+- **Pattern Recognition**: Advanced code pattern matching and similarity detection
+- **Comprehensive Coverage**: Both internal code exploration and external best practice lookup
+
+**Implementation Approach Planning**:
+Each task's `flow_control.implementation_approach` defines execution strategy (planning phase):
+
+1. **task_description**: Implementation strategy definition:
+   - Clear implementation goal to be executed later
+   - Planned reference to patterns from pre_analysis results
+   - Integration strategy with existing codebase
+
+2. **modification_points**: Planned code modification targets:
+   - Specific code changes to be made during execution
+   - Planned use of parent task patterns via `[parent]` context
+   - Integration points with existing components via `[context]` from dependencies
+
+3. **logic_flow**: Business logic execution plan:
+   - Step-by-step workflow to be implemented
+   - Planned data flow between components
+   - Integration points using `[inherited]` and `[shared]` context
+
+4. **target_files**: Target file specifications for execution:
+   - `src/auth/login.ts:handleLogin:75-120` (planned function and line range)
+   - `src/middleware/auth.ts:validateToken` (planned function target)
+   - Must align with task's `context.focus_paths`
+
+**Variable Reference System**:
+- `[design]` - Results from pre_analysis steps
+- `[parent]` - Context inherited from parent tasks
+- `[context]` - Dependencies from related tasks
+- `[inherited]` - Shared context from session
+- `[shared]` - Global rules and patterns 

 **Content Sources**:
 - Task summaries: `.workflow/WFS-[session]/.summaries/`
@@ -157,6 +222,7 @@ Flow_control design should follow these principles:
 - ✅ **Correct (Function-based)**: `IMPL-001: User authentication system` (models + routes + components + middleware + tests)
 - ❌ **Wrong (File/step-based)**: `IMPL-001: Create database model`, `IMPL-002: Create API endpoint`

+
 ## Document Generation

 **Workflow**: Identifier Creation → Folder Structure → IMPL_PLAN.md → .task/IMPL-NNN.json → TODO_LIST.md
--- a/.claude/workflows/mcp-tool-strategy.md
+++ b/.claude/workflows/mcp-tool-strategy.md
@@ -0,0 +1,168 @@
+# MCP Tool Strategy: Triggers & Workflows
+
+## ⚡ Triggering Mechanisms
+
+**Auto-Trigger Scenarios**:
+- User mentions "exa-code" or code-related queries → `mcp__exa__get_code_context_exa`
+- Need current web information → `mcp__exa__web_search_exa`
+- Finding code patterns/files → `mcp__code-index__search_code_advanced`
+- Locating specific files → `mcp__code-index__find_files`
+
+**Manual Trigger Rules**:
+- Complex API research → Exa Code Context
+- Architecture pattern discovery → Exa Code Context + Gemini analysis
+- Real-time information needs → Exa Web Search
+- Codebase exploration → Code Index tools first, then Gemini analysis
+
+## 🎯 Available MCP Tools
+
+### Exa Code Context (mcp__exa__get_code_context_exa)
+**Purpose**: Search and get relevant context for programming tasks
+**Strengths**: Highest quality context for libraries, SDKs, and APIs
+**Best For**: Code examples, API patterns, learning frameworks
+
+**Usage**:
+```bash
+mcp__exa__get_code_context_exa(
+  query="React useState hook examples",
+  tokensNum="dynamic"  # or 1000-50000
+)
+```
+
+**Examples**: "React useState", "Python pandas filtering", "Express.js middleware"
+
+### Exa Web Search (mcp__exa__web_search_exa)
+**Purpose**: Real-time web searches with content scraping
+**Best For**: Current information, research, recent solutions
+
+**Usage**:
+```bash
+mcp__exa__web_search_exa(
+  query="latest React 18 features",
+  numResults=5  # default: 5
+)
+```
+
+
+### Code Index Tools (mcp__code-index__)
+**核心方法**: `search_code_advanced`, `find_files`, `refresh_index`
+
+**核心搜索**:
+```bash
+mcp__code-index__search_code_advanced(pattern="function.*auth", file_pattern="*.ts")
+mcp__code-index__find_files(pattern="*.test.js")
+mcp__code-index__refresh_index()  # git操作后刷新
+```
+
+**实用场景**:
+- **查找代码**: `search_code_advanced(pattern="old.*API")`
+- **定位文件**: `find_files(pattern="src/**/*.tsx")`
+- **更新索引**: `refresh_index()` (git操作后)
+
+## 📊 Tool Selection Matrix
+
+| Task | MCP Tool | Use Case | Integration |
+|------|----------|----------|-------------|
+| **Code Context** | Exa Code | API examples, patterns | → Gemini analysis |
+| **Research** | Exa Web | Current info, trends | → Planning phase |
+| **Code Search** | Code Index | Pattern discovery, file location | → Gemini analysis |
+| **Navigation** | Code Index | File exploration, structure | → Architecture phase |
+
+## 🚀 Integration Patterns
+
+### Standard Workflow
+```bash
+# 1. Explore codebase structure
+mcp__code-index__find_files(pattern="*async*")
+mcp__code-index__search_code_advanced(pattern="async.*function", file_pattern="*.ts")
+
+# 2. Get external context
+mcp__exa__get_code_context_exa(query="TypeScript async patterns", tokensNum="dynamic")
+
+# 3. Analyze with Gemini
+cd "src/async" && ~/.claude/scripts/gemini-wrapper -p "
+PURPOSE: Understand async patterns
+CONTEXT: Code index results + Exa context + @{src/async/**/*}
+EXPECTED: Pattern analysis
+RULES: Focus on TypeScript best practices
+"
+
+# 4. Implement with Codex
+codex -C src/async --full-auto exec "Apply modern async patterns" -s danger-full-access
+```
+
+### Enhanced Planning
+1. **Explore codebase** with Code Index tools
+2. **Research** with Exa Web Search
+3. **Get code context** with Exa Code Context
+4. **Analyze** with Gemini
+5. **Architect** with Qwen
+6. **Implement** with Codex
+
+## 🔧 Best Practices
+
+### Code Index
+- **Search first** - Use before external tools for codebase exploration
+- **Refresh after git ops** - Keep index synchronized
+- **Pattern specificity** - Use precise regex patterns for better results
+- **File patterns** - Combine with glob patterns for targeted search
+
+### Exa Code Context
+- **Use "dynamic" tokens** for efficiency
+- **Be specific** - include technology stack
+- **MANDATORY** when user mentions exa-code or code queries
+
+### Exa Web Search
+- **Default 5 results** usually sufficient
+- **Use for current info** - supplement knowledge cutoff
+
+
+
+## 🎯 Common Scenarios
+
+### Learning New Technology
+```bash
+# Explore existing patterns + get examples + research + analyze
+mcp__code-index__search_code_advanced(pattern="router|routing", file_pattern="*.ts")
+mcp__exa__get_code_context_exa(query="Next.js 14 app router", tokensNum="dynamic")
+mcp__exa__web_search_exa(query="Next.js 14 best practices 2024", numResults=3)
+cd "src/app" && ~/.claude/scripts/gemini-wrapper -p "Learn Next.js patterns"
+```
+
+### Debugging
+```bash
+# Find similar patterns + solutions + fix
+mcp__code-index__search_code_advanced(pattern="similar.*error", file_pattern="*.ts")
+mcp__exa__get_code_context_exa(query="TypeScript generic constraints", tokensNum="dynamic")
+codex --full-auto exec "Fix TypeScript issues" -s danger-full-access
+```
+
+### Codebase Exploration
+```bash
+# Comprehensive codebase understanding workflow
+mcp__code-index__set_project_path(path="/current/project")  # 设置项目路径
+mcp__code-index__refresh_index()  # 刷新索引
+mcp__code-index__find_files(pattern="*auth*")  # Find auth-related files
+mcp__code-index__search_code_advanced(pattern="function.*auth", file_pattern="*.ts")  # Find auth functions
+mcp__code-index__get_file_summary(file_path="src/auth/index.ts")  # Understand structure
+cd "src/auth" && ~/.claude/scripts/gemini-wrapper -p "Analyze auth architecture"
+```
+
+### Project Setup Workflow
+```bash
+# 新项目初始化流程
+mcp__code-index__set_project_path(path="/path/to/new/project")
+mcp__code-index__get_settings_info()  # 确认设置
+mcp__code-index__refresh_index()  # 建立索引
+mcp__code-index__configure_file_watcher(enabled=true)  # 启用文件监控
+mcp__code-index__get_file_watcher_status()  # 确认监控状态
+```
+
+## ⚡ Performance Tips
+
+- **Code Index first** → explore codebase before external tools
+- **Use "dynamic" tokens** for Exa Code Context
+- **MCP first** → gather context before analysis
+- **Focus queries** - avoid overly broad searches
+- **Integrate selectively** - use relevant context only
+- **Refresh index** after major git operations
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -7,6 +7,7 @@ This document defines project-specific coding standards and development principl
 For all CLI tool usage, command syntax, and integration guidelines:
 - **Intelligent Context Strategy**: @~/.claude/workflows/intelligent-tools-strategy.md
 - **Context Search Commands**: @~/.claude/workflows/context-search-strategy.md
+- **MCP Tool Strategy**: @~/.claude/workflows/mcp-tool-strategy.md

 **Context Requirements**:
 - Identify 3+ existing similar patterns before implementation
--- a/README_CN.md
+++ b/README_CN.md
@@ -25,6 +25,32 @@
 - **✅ 计划验证系统**: 使用 Gemini/Codex 双重分析的执行前验证 (`/workflow:plan-verify`)
 - **🎯 JSON优先架构**: 具有原子会话管理的单一数据源
 - **💡 头脑风暴产物**: 多视角规划与综合和结构化文档生成
+- **🚀 MCP 工具集成**: 通过模型上下文协议实现增强的代码库分析和外部上下文获取
+
+### 🔧 MCP 工具增强 (NEW!)
+
+**通过 MCP (Model Context Protocol) 工具实现的高级代码库分析:**
+
+#### 必需的 MCP 服务器
+1. **Exa MCP Server** - 外部 API 模式和示例
+   - 仓库: https://github.com/exa-labs/exa-mcp-server
+   - 功能: `mcp__exa__get_code_context_exa()` - 获取外部最佳实践
+
+2. **Code Index MCP** - 内部代码库探索
+   - 仓库: https://github.com/johnhuang316/code-index-mcp
+   - 功能:
+     - `mcp__code-index__find_files()` - 文件模式匹配
+     - `mcp__code-index__search_code_advanced()` - 高级代码搜索
+
+#### MCP 集成优势
+- **📊 更快分析**: 直接代码库索引 vs 手动文件搜索
+- **🌐 外部上下文**: 真实世界的 API 模式和实现示例
+- **🔍 模式识别**: 高级代码模式匹配和相似性检测
+- **📈 全面覆盖**: 内部代码探索和外部最佳实践查找
+- **⚡ 自动回退**: 如果 MCP 工具不可用，工作流使用传统 bash/CLI 工具
+
+#### 安装和设置
+请安装这些 MCP 服务器以启用增强的代码库分析。工作流将在可用时自动使用它们。

 ---

@@ -509,6 +535,9 @@ 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
+- **🔧 MCP 服务器**: 推荐用于增强代码库分析
+  - **Exa MCP Server**: 外部 API 模式和最佳实践获取
+  - **Code Index MCP**: 高级内部代码库搜索和索引

 ## 配置