Integrates with Git repositories to provide project-aware memory management, tracking coding decisions and changes within version-controlled codebases.
Provides comprehensive metadata management and safety controls for JavaScript projects, including method-level permissions and automated changelog generation.
Referenced in decision-making examples for database technology choices, with the server capable of tracking and remembering architectural decisions including PostgreSQL adoption reasoning.
Built with TypeScript support and provides metadata management capabilities for TypeScript projects through AI-metadata headers and file tracking.
MCP Memory Server with Folder Mapping
A comprehensive MCP (Model Context Protocol) server that provides AI coding assistants with memory management, file approval tracking, changelog functionality, and automatic folder mapping.
🆕 New Features Added
📁 Folder Mapping System
Automatically generates _map.md
files for each folder in your project, providing quick overviews of:
- Classes, interfaces, functions, and types
- Method signatures and parameters
- File purposes and relationships
- Dependencies and test locations
- Risk assessments
This solves the problem of code discoverability - helping both AI agents and developers quickly understand what's available without diving into every file.
🚀 Quick Start
1. Install Dependencies
2. Test the System
3. Build the Project
4. Run the MCP Server
🛠️ Available MCP Tools
Memory Management
start_session(task)
- Start a new coding sessionadd_session_step(step, filesModified, description?)
- Record completed workadd_decision(key, value, reasoning)
- Record important decisionsget_project_memory()
- Get current project state
File Approvals
set_file_approval(filePath, approvalType, approvedBy)
- Set approval statusget_file_approval_status(filePath)
- Check approval statuscheck_before_modification(filePath)
- Validate before changesget_modification_actions(filePath)
- Get post-change actions
📁 Folder Mapping (NEW)
generate_folder_map(folderPath)
- Generate_map.md
for specific foldergenerate_all_folder_maps()
- Generate maps for all project foldersupdate_folder_map_if_needed(filePath)
- Update map if files changedfind_all_map_files()
- List all existing map filesvalidate_all_maps()
- Check if maps are up-to-date
Metadata & Changelog
parse_file_metadata(filePath)
- Extract AI metadata from filesupdate_file_metadata(filePath, updates)
- Update file metadataadd_changelog_entry(...)
- Add changelog entryget_recent_changes(days?)
- Get recent project changes
📋 Folder Map Example
When you run generate_folder_map("src/services")
, it creates src/services/_map.md
:
🔧 Integration with Your Test Project
For your Zendesk-ClickUp automation project:
- Navigate to your test project:
- Copy the MCP server files or install as dependency
- Generate folder maps:
- Your folder structure will get maps:
🛡️ Memory Recording Fix
The memory recording issues have been fixed by:
- Ensuring directory creation before all file operations
- Better error handling with detailed logging
- Automatic retry logic for file system operations
- Proper async/await patterns throughout
💡 Benefits for AI Agents
Before Folder Mapping:
- AI has to read every file to understand what's available
- Often creates duplicate functionality
- Slow to understand project structure
- Poor code reuse suggestions
After Folder Mapping:
- Instant project comprehension from
_map.md
files - Prevents duplicate code - AI sees what already exists
- Better suggestions - AI knows available methods and classes
- Faster development - Quick navigation and understanding
📖 Usage in AI Workflows
Typical AI Session:
🎯 Best Practices
For AI Agents:
- Always read relevant
_map.md
files before suggesting new code - Use
update_folder_map_if_needed()
after creating/modifying files - Check existing functionality to avoid duplication
For Developers:
- Review generated maps for accuracy
- Update maps when adding major new features
- Use maps for onboarding new team members
🔄 Auto-Update Strategy
Maps can be kept current by:
- File watchers - Update when source files change
- Build integration - Generate maps during build process
- Git hooks - Update maps on commits
- CI/CD integration - Validate maps in pipelines
🚨 Troubleshooting
Memory Not Recording:
Maps Not Generating:
MCP Connection Issues:
📚 Next Steps
- Test the system with your Zendesk-ClickUp project
- Generate initial maps to see the folder mapping in action
- Integrate with your AI workflow using the MCP tools
- Customize map templates if needed for your specific patterns
The folder mapping feature transforms how AI agents understand and work with your codebase, making them far more effective at code reuse and avoiding duplication.
local-only server
The server can only run on the client's local machine because it depends on local resources.
Tools
Gives AI coding assistants persistent memory, safety controls, and project awareness by tracking coding sessions, protecting critical files from modifications, and managing approval workflows with automatic changelog generation.
Related MCP Servers
- -securityAlicense-qualityProvides code manipulation, execution, and version control capabilities. It allows AI assistants to read, write, and execute code while maintaining a history of changes.Last updated -9MIT License
- -securityFlicense-qualityA personal AI coding assistant that connects to various development environments and helps automate tasks, provide codebase insights, and improve coding decisions by leveraging the Model Context Protocol.Last updated -
- AsecurityAlicenseAqualityAdds persistent task management to AI coding assistants in IDEs, allowing them to track multi-step tasks across sessions and maintain organized workflow with project-scoped todo lists.Last updated -220MIT License
- AsecurityAlicenseAqualityProvides AI assistants with persistent memory of your project architecture, development history, and technical decisions, allowing them to give context-aware coding help without needing repeated explanations.Last updated -16611MIT License