Skip to main content
Glama

DollhouseMCP

by DollhouseMCP
RELEASE_NOTES_v1.9.0.mdโ€ข4.26 kB
# Release Notes - v1.9.0 **Release Date**: September 19, 2025 **Type**: Minor Release **Codename**: Memory Milestone ## ๐ŸŽ‰ Highlights We're excited to announce v1.9.0 of DollhouseMCP, featuring the highly anticipated **Memory element type**! This release marks a major milestone with PR #1000, introducing persistent context storage with enterprise-grade scalability and security features. ## โœจ New Features ### Memory Element Implementation ๐Ÿง  The Memory element provides persistent context storage with advanced features: - **Smart Organization**: Automatic date-based folder structure (YYYY-MM-DD) prevents performance issues at scale - **Deduplication**: SHA-256 content hashing prevents duplicate storage - **Fast Search**: Built-in indexing for queries across thousands of entries - **Privacy Controls**: Three-tier privacy levels (private, team, public) - **Retention Policies**: Automatic cleanup based on age and capacity - **Tag System**: Organize and find memories with flexible tagging - **Rich Metadata**: Track statistics, timestamps, and context ### Performance Enhancements โšก - 60-second cache for date folder operations - Atomic file operations prevent corruption - Efficient collision handling with version suffixes - Optimized search indexing for large datasets ### Security Improvements ๐Ÿ”’ - Comprehensive input validation and sanitization - Security event audit logging throughout - Memory size limits to prevent DoS attacks - Content hash verification for integrity ## ๐Ÿ“Š Technical Details ### Memory Storage Structure ``` ~/.dollhouse/portfolio/memories/ โ”œโ”€โ”€ 2025-09-18/ โ”‚ โ””โ”€โ”€ project-context.yaml โ”œโ”€โ”€ 2025-09-19/ โ”‚ โ”œโ”€โ”€ meeting-notes.yaml โ”‚ โ””โ”€โ”€ code-review.yaml ``` ### Key Capabilities - Add unlimited entries per memory (with configurable limits) - Search by content, tags, or metadata - Automatic retention policy enforcement - Statistics tracking (size, count, dates) - Multiple storage backends (file system now, database support planned) ## ๐Ÿ› Bug Fixes - Fixed CI test conflicts with GitHub integration tests (#1001) - Resolved race conditions in concurrent file operations - Fixed Unicode normalization issues in memory content ## ๐Ÿ”„ Changes - Memory files now save to date-based folders instead of flat directory - Content deduplication enabled by default - Search indexing triggers automatically at 100+ entries ## โš ๏ธ Known Issues - Prototype pollution warnings in ConfigManager (low priority, main branch only) - Memory encryption not yet implemented (planned for v2.0) ## ๐Ÿ“ˆ Statistics - **89 tests** for memory functionality alone - **100% pass rate** on all CI workflows - **2 major PRs** merged (#1000, #1001) - **3+ issues** resolved or advanced ## ๐Ÿš€ Getting Started with Memories ### Create a Memory ```typescript import { Memory } from '@dollhousemcp/mcp-server'; const memory = new Memory({ name: 'project-context', description: 'Context for my project', tags: ['project', 'important'] }); // Add entries await memory.addEntry('Design decision: Use TypeScript', ['decision', 'architecture']); await memory.addEntry('Meeting notes from standup', ['meeting', 'daily']); ``` ### Search Memories ```typescript // Search by content const results = await memory.search({ query: 'TypeScript' }); // Search by tags const meetings = await memory.search({ tags: ['meeting'] }); ``` ## ๐Ÿ’ป Installation ```bash npm install @dollhousemcp/mcp-server@1.9.0 ``` ## ๐Ÿ“‹ Migration No migration needed - Memories are a new feature with no existing data. ## ๐Ÿ™ Acknowledgments Special thanks to everyone who contributed to this milestone release, especially reaching PR #1000! ## ๐Ÿ“ Full Changelog For a complete list of changes, see the [full changelog](https://github.com/DollhouseMCP/mcp-server/compare/v1.8.1...v1.9.0). ## ๐Ÿ”ฎ What's Next - **v2.0.0**: Memory encryption and cloud sync - **v1.10.0**: Enhanced search with fuzzy matching - **v1.11.0**: Memory templates and presets --- **Note**: This is an MVP release of the Memory feature. We welcome feedback and bug reports at our [GitHub repository](https://github.com/DollhouseMCP/mcp-server/issues). ๐Ÿค– Generated with [Claude Code](https://claude.ai/code)

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/DollhouseMCP/DollhouseMCP'

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