mirror of
https://github.com/catlog22/Claude-Code-Workflow.git
synced 2026-02-10 02:24:35 +08:00
## 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>
4.3 KiB
4.3 KiB
name, description, usage, examples
| name | description | usage | examples | |
|---|---|---|---|---|
| update-memory-full | Complete project-wide CLAUDE.md documentation update | /update-memory-full |
|
🚀 Command Overview: /update-memory-full
Complete project-wide documentation update using depth-parallel execution.
📝 Execution Template
#!/bin/bash
# Complete project-wide CLAUDE.md documentation update
# 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 3: Get and display project structure
modules=$(Bash(~/.claude/scripts/get_modules_by_depth.sh list))
count=$(echo "$modules" | wc -l)
# Step 3: Analysis handover → Model takes control
# BASH_EXECUTION_STOPS → MODEL_ANALYSIS_BEGINS
# Pseudocode flow:
# IF (module_count > 20 OR complex_project_detected):
# → Task "Complex project full update" subagent_type: "memory-gemini-bridge"
# ELSE:
# → Present plan and WAIT FOR USER APPROVAL before execution
🧠 Model Analysis Phase
After the bash script completes the initial analysis, the model takes control to:
- Analyze Complexity: Review module count and project context
- Parse CLAUDE.md Status: Extract which modules have/need CLAUDE.md files
- Choose Strategy:
- Simple projects: Present execution plan with CLAUDE.md paths to user
- Complex projects: Delegate to memory-gemini-bridge agent
- Present Detailed Plan: Show user exactly which CLAUDE.md files will be created/updated
- ⚠️ CRITICAL: WAIT FOR USER APPROVAL: No execution without explicit user consent
📋 Execution Strategies
For Simple Projects (≤20 modules):
Model will present detailed plan:
📋 Update Plan:
NEW CLAUDE.md files (X):
- ./src/components/CLAUDE.md
- ./src/services/CLAUDE.md
UPDATE existing CLAUDE.md files (Y):
- ./CLAUDE.md
- ./src/CLAUDE.md
- ./tests/CLAUDE.md
Total: N CLAUDE.md files will be processed
⚠️ Confirm execution? (y/n)
# ⚠️ MANDATORY: User confirmation → MUST await explicit approval
IF user_explicitly_confirms():
continue_execution()
ELSE:
abort_execution()
# Step 4: Organize modules by depth → Prepare execution
depth_modules = organize_by_depth(modules)
# Step 5: Execute depth-parallel updates → Process by depth
FOR depth FROM max_depth DOWN TO 0:
FOR each module IN depth_modules[depth]:
WHILE active_jobs >= 4: wait(0.1)
Bash(~/.claude/scripts/update_module_claude.sh "$module" "full" &)
wait_all_jobs()
# Step 6: Safety check and restore staging state
non_claude=$(Bash(git diff --cached --name-only | grep -v "CLAUDE.md" || true))
if [ -n "$non_claude" ]; then
Bash(git restore --staged .)
echo "⚠️ Warning: Non-CLAUDE.md files were modified, staging reverted"
echo "Modified files: $non_claude"
else
echo "✅ Only CLAUDE.md files modified, staging preserved"
fi
# Step 7: Display changes → Final status
Bash(git status --short)
For Complex Projects (>20 modules): The model will delegate to the memory-gemini-bridge agent using the Task tool:
Task "Complex project full update"
prompt: "Execute full documentation update for [count] modules using depth-parallel execution"
subagent_type: "memory-gemini-bridge"
📚 Usage Examples
# Complete project documentation refresh
/update-memory-full
# After major architectural changes
/update-memory-full
✨ Features
- Separated Commands: Each bash operation is a discrete, trackable step
- Intelligent Complexity Detection: Model analyzes project context for optimal strategy
- Depth-Parallel Execution: Same depth modules run in parallel, depths run sequentially
- Git Integration: Auto-cache changes before, safety check and show status after
- Safety Protection: Automatic detection and revert of unintended source code modifications
- Module Detection: Uses get_modules_by_depth.sh for structure discovery
- User Confirmation: Clear plan presentation with approval step
- CLAUDE.md Only: Only updates documentation, never source code