Skip to main content
Glama

mcp-adr-analysis-server

by tosin2013
MIGRATION_GUIDE.md5.46 kB
# Migration Guide for Deprecated Tools ## Overview This guide helps you migrate from deprecated tools to the new memory-centric architecture in MCP ADR Analysis Server v2.0+. ## Deprecated Tools and Their Replacements ### 1. `mcp0_manage_todo_json` → `mcp-shrimp-task-manager` **What Changed:** - The built-in TODO management has been replaced with an external task manager - Task data is now stored in the memory system rather than JSON files **Migration Steps:** 1. Export existing tasks using the old tool (if still available) 2. Install the new task manager: `mcp-shrimp-task-manager` 3. Import tasks into the new system 4. Update any automation scripts to use the new tool names **New Tool Usage:** ```javascript // Old way try { await mcp0_manage_todo_json({ operation: 'add', task: 'My task' }); } catch (error) { console.error('Failed to manage todo:', error); } // New way (via external task manager) try { // Configure mcp-shrimp-task-manager in your MCP settings // Then use through MCP client const result = await mcpClient.callTool('manage-task', { operation: 'add', task: 'My task' }); console.log('Task added successfully:', result); } catch (error) { console.error('Failed to add task:', error); } ``` ### 2. `mcp0_smart_score` → Memory-Centric Health Scoring **What Changed:** - Project health scoring is now integrated into the memory system - Provides more comprehensive, context-aware scoring - Historical tracking and pattern recognition **Migration Steps:** 1. The new system automatically migrates existing score data 2. Access scores through memory entities instead of direct tool calls **New Usage:** ```javascript // Old way try { const score = await mcp0_smart_score({ projectPath: '/my/project' }); console.log('Project score:', score); } catch (error) { console.error('Failed to get project score:', error); } // New way (automatic via memory system) try { // Scores are calculated automatically and stored as memory entities const memoryQuery = await mcpClient.callTool('query-memory', { entityType: 'project-health', projectPath: '/my/project' }); console.log('Health data:', memoryQuery.entities); } catch (error) { console.error('Failed to query memory system:', error); } ``` ## Cache Directory Changes ### Important: New Cache Location **Old Location:** `.mcp-adr-cache` (in project directory) **New Location:** OS temp directory (e.g., `/tmp/[project-name]/cache`) This change: - Prevents permission issues - Provides better cross-platform compatibility - Keeps project directories clean **No Action Required:** The system automatically uses the new location. ## Memory System Integration ### New Features The memory-centric architecture provides: - **Persistent memory entities** for all tool operations - **Cross-tool relationships** and pattern detection - **Historical analysis** and trend identification - **Intelligent caching** with automatic invalidation ### How to Enable Memory features are enabled by default. To use advanced features: ```javascript // Enable memory integration in tools try { const result = await deploymentReadiness({ enableMemoryIntegration: true, migrateExistingHistory: true, }); console.log('Memory integration enabled:', result); } catch (error) { console.error('Failed to enable memory integration:', error); // Fallback to basic mode console.log('Continuing with basic deployment readiness...'); } ``` ## Troubleshooting ### Common Issues 1. **"Tool not found" errors** - Solution: Update your MCP client configuration - Remove references to deprecated tools - Add new external tools as needed 2. **Cache directory errors** - Solution: Already fixed! Uses OS temp directory - Clear old cache: `rm -rf .mcp-adr-cache` 3. **Memory initialization failures** - Solution: Ensure temp directory has write permissions - Check disk space in temp directory ### Getting Help - Report issues: https://github.com/anthropics/claude-code/issues - Check documentation: `/docs` directory - Memory system guide: `/./explanation/memory-architecture.md` ## Rollback Instructions If you need to rollback to previous versions: 1. **Backup current data:** ```bash cp -r /tmp/[project-name]/cache ./backup-cache cp -r /tmp/[project-name]/memory ./backup-memory ``` 2. **Downgrade package:** ```bash npm install mcp-adr-analysis-server@1.x ``` 3. **Restore old cache location:** ```bash cp -r ./backup-cache .mcp-adr-cache ``` ## Benefits of Migration ### Why Migrate? 1. **Better Performance** - Memory-centric caching reduces redundant operations - Pattern recognition improves over time 2. **Enhanced Intelligence** - Cross-tool learning and relationships - Historical pattern analysis - Predictive capabilities 3. **Improved Reliability** - No more permission issues - Better error handling and recovery - Automatic data migration 4. **Future-Proof** - Active development and support - New features added regularly - Community-driven improvements ## Timeline - **Immediate:** Cache directory fixes are live - **v2.1:** Full memory system integration - **v2.2:** Advanced pattern recognition - **v3.0:** AI-powered workflow automation ## Questions? For questions about migration, please: 1. Check this guide first 2. Review the changelog in `/CHANGELOG.md` 3. Open an issue if you encounter problems

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/tosin2013/mcp-adr-analysis-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server